100% found this document useful (2 votes)
31 views

PDF Functional Protein Microarrays in Drug Discovery 1st Edition Paul F. Predki download

Protein

Uploaded by

yottaelev
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
100% found this document useful (2 votes)
31 views

PDF Functional Protein Microarrays in Drug Discovery 1st Edition Paul F. Predki download

Protein

Uploaded by

yottaelev
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 81

Download the full version of the ebook at

https://ebookultra.com

Functional Protein Microarrays in Drug


Discovery 1st Edition Paul F. Predki

https://ebookultra.com/download/functional-
protein-microarrays-in-drug-discovery-1st-edition-
paul-f-predki/

Explore and download more ebook at https://ebookultra.com


Recommended digital products (PDF, EPUB, MOBI) that
you can download immediately if you are interested.

Protein Discovery Technologies Drug Discovery Series 1st


Edition Renata Pasqualini

https://ebookultra.com/download/protein-discovery-technologies-drug-
discovery-series-1st-edition-renata-pasqualini/

ebookultra.com

Protein Crystallography in Drug Discovery Methods and


Principles in Medicinal Chemistry 1st Edition Robert E.
Babine
https://ebookultra.com/download/protein-crystallography-in-drug-
discovery-methods-and-principles-in-medicinal-chemistry-1st-edition-
robert-e-babine/
ebookultra.com

Phage Display In Biotechnology and Drug Discovery Drug


Discovery Series 1st Edition Sachdev S. Sidhu

https://ebookultra.com/download/phage-display-in-biotechnology-and-
drug-discovery-drug-discovery-series-1st-edition-sachdev-s-sidhu/

ebookultra.com

Protein Microarrays Methods and Protocols 2011th Edition


Ulrike Korf

https://ebookultra.com/download/protein-microarrays-methods-and-
protocols-2011th-edition-ulrike-korf/

ebookultra.com
Optimization in Drug Discovery 1st Edition Yan Z.

https://ebookultra.com/download/optimization-in-drug-discovery-1st-
edition-yan-z/

ebookultra.com

Accounts in Drug Discovery Case Studies in Medicinal


Chemistry RSC Drug Discovery Series Volume 4 1st Edition.
Edition Joel C. Barrish
https://ebookultra.com/download/accounts-in-drug-discovery-case-
studies-in-medicinal-chemistry-rsc-drug-discovery-series-volume-4-1st-
edition-edition-joel-c-barrish/
ebookultra.com

Drug Discovery in Cancer Epigenetics 1st Edition Gerda


Egger

https://ebookultra.com/download/drug-discovery-in-cancer-
epigenetics-1st-edition-gerda-egger/

ebookultra.com

Target Validation in Drug Discovery 1st Edition Brian W.


Metcalf

https://ebookultra.com/download/target-validation-in-drug-
discovery-1st-edition-brian-w-metcalf/

ebookultra.com

Pseudo peptides in Drug Discovery 1st Edition Chi-Huey


Wong

https://ebookultra.com/download/pseudo-peptides-in-drug-discovery-1st-
edition-chi-huey-wong/

ebookultra.com
Functional Protein Microarrays in Drug Discovery 1st
Edition Paul F. Predki Digital Instant Download
Author(s): Paul F. Predki
ISBN(s): 9780849398094, 0849398096
Edition: 1
File Details: PDF, 14.73 MB
Year: 2007
Language: english
9809_C000.fm Page i Saturday, February 17, 2007 1:53 PM

Functional Protein
Microarrays in
Drug Discovery
9809_C000.fm Page ii Saturday, February 17, 2007 1:53 PM

Drug Discovery Series

Series Editor

Andrew Carmen
Johnson & Johnson PRD, LLC
San Diego, California, U.S.A.

1. Virtual Screening in Drug Discovery, edited by Juan Alvarez


and Brian Shoichet
2. Industrialization of Drug Discovery: From Target Selection Through
Lead Optimization, edited by Jeffrey S. Handen, Ph.D.
3. Phage Display in Biotechnology and Drug Discovery, edited by
Sachdev S. Sidhu
4. G Protein-Coupled Receptors in Drug Discovery, edited by
Kenneth H. Lundstrom and Mark L. Chiu
5. Handbook of Assay Development in Drug Discovery, edited by
Lisa K. Minor
6. In Silico Technologies in Drug Target Identification and Validation,
edited by Darryl León and Scott Markel
7. Biochips as Pathways to Drug Discovery, edited by Andrew Carmen
and Gary Hardiman
8. Functional Protein Microarrays in Drug Discovery, edited by
Paul F. Predki
9809_C000.fm Page iii Saturday, February 17, 2007 1:53 PM

Functional Protein
Microarrays in
Drug Discovery
edited by
Paul F. Predki

Boca Raton London New York

CRC Press is an imprint of the


Taylor & Francis Group, an informa business
9809_C000.fm Page iv Saturday, February 17, 2007 1:53 PM

CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742
© 2007 by Taylor & Francis Group, LLC
CRC Press is an imprint of Taylor & Francis Group, an Informa business

No claim to original U.S. Government works


Printed in the United States of America on acid-free paper
10 9 8 7 6 5 4 3 2 1

International Standard Book Number-10: 0-8493-9809-6 (Hardcover)


International Standard Book Number-13: 978-0-8493-9809-4 (Hardcover)

This book contains information obtained from authentic and highly regarded sources. Reprinted
material is quoted with permission, and sources are indicated. A wide variety of references are
listed. Reasonable efforts have been made to publish reliable data and information, but the author
and the publisher cannot assume responsibility for the validity of all materials or for the conse-
quences of their use.

No part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any
electronic, mechanical, or other means, now known or hereafter invented, including photocopying,
microfilming, and recording, or in any information storage or retrieval system, without written
permission from the publishers.

For permission to photocopy or use material electronically from this work, please access www.
copyright.com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC)
222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that
provides licenses and registration for a variety of users. For organizations that have been granted a
photocopy license by the CCC, a separate system of payment has been arranged.

Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and
are used only for identification and explanation without intent to infringe.

Library of Congress Cataloging-in-Publication Data

Functional protein microarrays in drug discovery / [edited by] Paul Predki.


p. ; cm. -- (Drug discovery series ; 8)
Includes bibliographical references and index.
ISBN-13: 978-0-8493-9809-4 (hardcover : alk. paper)
ISBN-10: 0-8493-9809-6 (hardcover : alk. paper)
1. Protein microarrays. 2. Drugs--Design. I. Predki, Paul. II. Series.
[DNLM: 1. Proteins--analysis. 2. Drug Design. 3. Protein Array Analysis. QU
55 F965 2007]

QP551.F96 2007
572’.636--dc22 2007005653

Visit the Taylor & Francis Web site at


http://www.taylorandfrancis.com
and the CRC Press Web site at
http://www.crcpress.com
9809_C000.fm Page v Saturday, February 17, 2007 1:53 PM

Table of Contents
Section 1
Functional Protein Content for Microarrays .......................................................1

Chapter 1 High-Throughput Gene Cloning Using the Gateway®


Technology .......................................................................................... 3
Scott N. Peterson, Patrick Burr, Getahun Tsegaye, and Pratap Venepally

Chapter 2 Protein Expression for MicroArrays..................................................23


Harry H. Yim, Thomas G. Chappell, and Steven H. Harwood

Chapter 3 Emerging Trend: Cell-Free Protein Expression ................................39


Federico Katzen and Wieslaw Kudlicki

Section 2
Fabrication of Functional Protein Microarrays.................................................51

Chapter 4 The Critical Role of Surface Chemistry


in Protein Microarrays .......................................................................53
Athena Guo and X.-Y. Zhu

Chapter 5 Fabrication of Sol-Gel-Derived Protein Microarrays


for Diagnostics and Screening...........................................................73
Nicholas Rupcich and John D. Brennan

Chapter 6 Printing and QC of Functional Protein Microarrays.........................99


Dee Shen, Fang X. Zhou, and Barry Schweitzer

Chapter 7 Protein Engineering for Surface Attachment...................................115


Aparna Girish, Grace Y. J. Chen, and Shao Q. Yao

Chapter 8 Protein In Situ Arrays through Cell-Free Protein Synthesis ...........133


Mingyue He, Farid Khan, Elizabeth Palmer, Mingwei Wang,
and Michael J. Taussig
9809_C000.fm Page vi Saturday, February 17, 2007 1:53 PM

Section 3
Detection Methods for Protein Microarrays ....................................................145

Chapter 9 Fluorescent Detection Methods for Protein Microarrays............... 147


Steven Roman and Scott Clarke

Chapter 10 Functional Analysis of Protein Interactions Using Surface


Plasmon Resonance-Based Microarrays..........................................181
Alan McWhirter and Stefan Löfås

Chapter 11 Leaving the Surface Behind: At the Intersection of Protein


Microarrays and Mass Spectrometry...............................................199
Darrell P. Chandler, Daniel S. Schabacker, Sergei Bavykin,
and Igor M. Gavin

Chapter 12 High-Resolution Label-Free Detection Applied to


Protein Microarray Research ...........................................................217
Lance G. Laing and Brian Cunningham

Section 4
Applications of Functional Protein Microarrays .............................................237

Chapter 13 Studying Protein–Protein Interactions with Protein Microarrays:


Rapid Identification of 14-3-3 Protein Binding Partners ................239
Jun-ichi Satoh

Chapter 14 A Combined Force of Chemical Genetics


and Protein Microarrays...................................................................261
Heng Zhu and Jing Huang

Chapter 15 Antibody Profiling for Protein Drug Development


and Clinical Development................................................................275
Steve H. Herrmann

Chapter 16 Humoral Response Profiling Using Protein Microarrays ...............301


Arun Sreekumar, Barry S. Taylor, Xiaoju Wang, David Lubman,
and Arul M. Chinnaiyan
9809_C000.fm Page vii Saturday, February 17, 2007 1:53 PM

Chapter 17 DNA Interactions with Arrayed Proteins ........................................313


Marina Snapyan and Vehary Sakanyan

Chapter 18 G Protein–Coupled Receptor Microarrays for Drug Discovery .....333


John Salon, Michael Johnson, Brian Rasnow, Gloria Biddlecome, Yulong Hong,
Brian Webb, Ye Fang, and Joydeep Lahiri

Chapter 19 Kinase Substrate Identification Using Yeast


Protein Microarrays..........................................................................351
Geeta Devgan and Michael Snyder

Section 5
Bioinformatics & Data Analysis.........................................................................361

Chapter 20 Protein Microarray Image Analysis.................................................363


Minzi Ruan

Chapter 21 The Analysis of Protein Arrays .......................................................381


Brad Love

Chapter 22 Evaluating Precision and Recall in Functional


Protein Arrays ..................................................................................403
Keith Robison

Chapter 23 Visualization of Protein Microarray Data .......................................415


Kevin Clancy

Index......................................................................................................................443
9809_C000.fm Page viii Saturday, February 17, 2007 1:53 PM
9809_C000.fm Page ix Saturday, February 17, 2007 1:53 PM

Preface
Since their introduction in the 1990s, microarray-based technologies have had a
tremendous impact on the biological sciences. One of the most exciting recent
developments in this field is functional protein microarrays: microarrays with large
numbers of correctly folded and functional proteins. Initially considered an imprac-
tical if not impossible goal, high-content functional protein microarrays have now
proven their utility in a multitude of applications. While the ‘‘field’s” early successes
have set the stage for the rapid growth now being witnessed, it is not without its
challenges. Indeed, challenges are to be expected in a fast-moving interdisciplinary
endeavor such as this, where molecular biology, protein chemistry, bioinformatics,
engineering, and physical sciences all intersect.
Currently no book has addressed all aspects of functional protein microarrays
in a coherent and integrated fashion. This book is intended to provide the first
comprehensive reference for the field, addressing basic principles, methods, and
applications. While intended primarily as a reference for industrial, academic, and
government scientists, it is also suitable as a graduate-level supplementary text. The
book is divided into five main sections, each addressing critical aspects of the field.
The first focuses on the generation of functional protein content, which is the first
and perhaps most challenging aspect of protein microarrays. The second section
describes both “standard” and state-of-the-art fabrication methods, focusing on
issues of particular significance to functional protein microarrays. Similarly, the third
section reviews current and next-generation approaches to assay detection, which
hold one key to the future of the field. The fourth and largest section is dedicated
to applications. This section spans the breadth of published applications, from
biomolecular interaction discovery and characterization (proteins, antibodies, DNA,
small molecules) to humoral response biomarker profiling, enzyme substrate iden-
tification, and drug discovery. The final section addresses fundamental computational
issues including image and data analysis as well as data visualization.
The intent of this book is to provide the first integrated reference for functional
protein microarrays. In doing so, I have aspired to create a volume worthy of the
promise of functional protein microarrays, a practical resource capable of conveying
the excitement and enabling the development of this field. This book would not have
been possible, however, without the hard work of its many authors and Kathie
McCoy, to whom I am truly grateful.
9809_C000.fm Page x Saturday, February 17, 2007 1:53 PM
9809_C000.fm Page xi Saturday, February 17, 2007 1:53 PM

Contributors List
Sergei Bavykin Arul M. Chinnaiyan
Argonne National Laboratory Department of Pathology
Argonne, Illinois and
Department of Urology
Gloria Biddlecome and
Amgen Incorporated Bioinformatics Program
Thousand Oaks, California and
Comprehensive Cancer Center
John D. Brennan University of Michigan Medical School
Department of Chemistry Ann Arbor, Michigan
McMaster University
Hamilton, Ontario Kevin Clancy
Informax
Patrick Burr Frederick, Maryland
The Institute for Genomic Research
The Pathogen Functional Genomics Scott Clarke
Molecular Probes
Resource Center
Eugene, Oregon
Rockville, Maryland
Brian Cunningham
Darrell P. Chandler
Department of Electrical and
Argonne National Laboratory
Computer Engineering
Argonne, Illinois
Micro and Nanotechnology Laboratory
and
University of Illinois at
Akonni Biosystems, Incorporated
Urbana-Champaign
Frederick, Maryland
Urbana, Illinois

Thomas G. Chappell Geeta Devgan


Invitrogen Corporation Department of Molecular, Cellular and
Carlsbad, California Developmental Biology
Yale University
Grace Y. J. Chen New Haven, Connecticut
Departments of Chemistry and
Biological Sciences Ye Fang
Medicinal Chemistry Program of Biochemical Technologies, Science and
the Office of Life Sciences Technology Division
National University of Singapore Corning Incorporated
Republic of Singapore Corning, New York
9809_C000.fm Page xii Saturday, February 17, 2007 1:53 PM

Igor M. Gavin Federico Katzen


Argonne National Laboratory Invitrogen Corporation
Argonne, Ilinois Carlsbad, California
and
University of Illinois at Chicago Farid Khan
Chicago, Illinois Technology Research Group
Protein Technologies Laboratory
Aparna Girish The Babraham Institute
Departments of Chemistry and Cambridge, U.K.
Biological Sciences
Medicinal Chemistry Program of Wieslaw Kudlicki
the Office of Life Sciences Invitrogen Corporation
National University of Singapore Carlsbad, California
Republic of Singapore
Joydeep Lahiri
Athena Guo Biochemical Technologies, Science
MicroSurfaces, Inc. and Technology Division
Minneapolis, Minnesota Corning Incorporated
Corning, New York
Steven H. Harwood
Eugene, Oregon Lance G. Laing
SRU Biosystems
Mingyue He
Woburn, Massachusetts
Technology Research Group
Protein Technologies Laboratory
Stefan Löfås
The Babraham Institute
Biacore AB, Rapsgatan
Cambridge, U.K.
Uppsala, Sweden
Steve H. Herrmann Brad Love
Biological Technologies Corporate Research Laboratory
Wyeth Research Invitrogen Corporation
Cambridge, Massachusetts Carlsbad, California
Yulong Hong David Lubman
Biochemical Technologies, Science Department of Chemistry
and Technology Division and
Corning Incorporated Department of Surgery
Corning, New York University of Michigan Medical School
Ann Arbor, Michigan
Jing Huang
Department of Molecular and Medical Alan McWhirter
Pharmacology Biacore AB, Rapsgatan
David Geffen School of Medicine Uppsala, Sweden
University of California
Los Angeles, California Elizabeth Palmer
Technology Research Group
Michael Johnson Protein Technologies Laboratory
Amgen Incorporated The Babraham Institute
Thousand Oaks, California Cambridge, U.K.
9809_C000.fm Page xiii Saturday, February 17, 2007 1:53 PM

Scott N. Peterson Daniel S. Schabacker


The Institute for Genomic Research Argonne National Laboratory
The Pathogen Functional Genomics Argonne, Illinois
Resource Center
Rockville, Maryland Barry Schweitzer
Protein Array Center
Brian Rasnow Invitrogen Corporation
Amgen Incorporated Branford, Connecticut
Thousand Oaks, California
Dee Shen
Keith Robison Protein Array Center
Computational Biology Invitrogen Corporation
Millennium Pharmaceuticals Branford, Connecticut
Incorporated
Cambridge, Massachusetts Marina Snapyan
Centre de Recherché Université Laval
Steven Roman Robert-Giffard (CRULRG)
Invitrogen Corporation Quebec, Quebec
Carlsbad, California
Michael Snyder
Minzi Ruan Department of Molecular, Cellular and
VigeneTech Incorporated Developmental Biology
Boston, Massachusetts Yale University
New Haven, Connecticut
Nicholas Rupcich
Department of Chemistry Arun Sreekumar
McMaster University Department of Pathology
Hamilton, Ontario and
Comprehensive Cancer Center
Vehary Sakanyan University of Michigan Medical School
Unité Biotechnologie, Biocatalyse, Ann Arbor, Michigan
Biorégulation
Université de Nantes
and Michael J. Taussig
ProtNeteomix Technology Research Group
Nantes, France Protein Technologies Laboratory,
The Babraham Institute
John Salon Cambridge, U.K.
Amgen Incorporated
Thousand Oaks, California Barry S. Taylor
Department of Pathology
Jun-ichi Satoh and
Department of Immunology Bioinformatics Program
National Institute of Neuroscience University of Michigan Medical School
Tokyo, Japan Ann Arbor, Michigan
9809_C000.fm Page xiv Saturday, February 17, 2007 1:53 PM

Getahun Tsegaye Shao Q. Yao


The Institute for Genomic Research Departments of Chemistry and
The Pathogen Functional Genomics Biological Sciences
Resource Center Medicinal Chemistry Program of the
Rockville, Maryland Office of Life Sciences
National University of Singapore
Pratap Venepally Republic of Singapore
The Institute for Genomic Research
The Pathogen Functional Genomics Harry H. Yim
Resource Center Invitrogen Corporation
Rockville, Maryland Carlsbad, California

Fang X. Zhou
Mingwei Wang
Protein Array Center
National Center for Drug Screening
Invitrogen Corporation
Shanghai Institute of Materia Medica
Branford, Connecticut
Shanghai, China
Heng Zhu
Xiaoju Wang Department of Pharmacology and
Department of Pathology HiT Center
University of Michigan Medical School The Johns Hopkins University School
Ann Arbor, Michigan of Medicine
Baltimore, Maryland
Brian Webb
Biochemical Technologies, Science X.-Y. Zhu
and Technology Division Department of Chemistry
Corning Incorporated University of Minnesota
Corning, New York Minneapolis, Minnesota
9809_C000.fm Page xv Saturday, February 17, 2007 1:53 PM

Introduction
As central actors in most biological responses, proteins are the subject of intense
study for both basic and drug research. This, in turn, has driven the development of
increasingly sophisticated approaches for the study of proteins, which, in recent
years has extended to proteomic level methodologies. Despite this need, however,
microarray technologies for proteins have lagged behind those for nucleic acids.
This has been particularly evident in the case of functional protein microarrays,
where formidable technical challenges must be surmounted. However, while tech-
nical challenges still remain, the past few years have witnessed movement of the
field from basic proof of concept1,2 to the use of microarrays for important scientific
work, landmark discoveries3 to proteomic characterizations.4 At the same time, the
number of publications in the field has increased exponentially. The purpose of this
book is to provide the reader with an up-to-date overview of the field, as well as
the background required to actually design and develop arrays or perform and
analyze array experiments. The five sections of this book reflect five key considera-
tions in the field: protein content, array fabrication, assay detection, applications and
data analysis.

FUNCTIONAL PROTEIN CONTENT


The development of functional protein content is one of the most challenging, and
often rate-limiting, aspects of protein microarray experimentation. These challenges
can be largely eliminated in cases where protein content or even protein arrays can
be acquired commercially. However, in many cases protein content must be generated
by the investigator. This content is most typically generated from DNA clones using
recombinant expression technology. High-throughput methods for expression clone
generation have been developed at The Institute for Genomic Research, and are
described in detail in Chapter 1. Important considerations such as information
management, automation, quality control and clone validation are addressed. The
second chapter addresses expression and purification of proteins in heterologous host
systems (E. coli, yeast and insect cells), and provides guidance for selecting an
appropriate system based on a variety of parameters such as yield, functionality,
post-translational modifications, throughput and cost. The final chapter of this section
reviews cell-free protein expression systems, and discusses specific considerations
for protein microarrays. Together, these chapters provide a thorough overview of
the basic considerations for protein content generation.
9809_C000.fm Page xvi Saturday, February 17, 2007 1:53 PM

FABRICATION
The functional and structural heterogeneity of proteins makes arraying and functional
surface attachment a considerable challenge. Chapter 4 provides a thorough exam-
ination of the challenges of surface chemistry for protein microarrays, which include
minimizing nonspecific interactions and maximizing the presentation of conforma-
tionally correct proteins. A completely different approach is described in Chapter 5
with the entrapment of proteins in a three-dimensional sol-gel. The various strategies
are illustrated in Figure 1.
Critical aspects of array manufacture are addressed in Chapter 6. These include
a brief review of commercially available printing technologies, the myriad challenges
presented by protein microarrays, and quality control in manufacturing.
The final chapters of this section describe novel strategies for generating
protein arrays. Chapter 7 focuses on oriented immobilization strategies based
on protein engineering and chemistry, while Chapter 8 addresses the in situ
generation of proteins. Both chapters describe methods that can “compress” the
steps involved in making an array, by combining purification (Chapter 7) or
expression and purification (Chapter 8) into the array printing process. These
simplified techniques promise to make protein microarray technology more
accessible to “average” labs, although at the potential cost of less well controlled
array content.

DETECTION
The varied applications of functional protein microarrays all require sensitive assay
detection technologies. The most common detection method, fluorescence, is
described in chapter 9. This chapter provides a detailed discussion of the basics:
fluorescent dyes, fluorescent proteins, time-resolved fluorescence, fluorescent quan-
tum dotes, signal amplification, labeling methods and instrumentation. It concludes

(a) (b) (c)

FIGURE 1 Protein immobilization strategies. (a) Proteins are directly attached to the surface
based on one (or few) site-specific interactions. This approach has the advantage of a relatively
homogeneous presentation of protein to the solution, but regions of the protein may be
systematically “hidden.” (b) Proteins are attached in a nonspecific orientation. This approach
has the advantage of (collectively) displaying a large fraction of the protein surface, but some
protein molecules may be functionally blocked. (c) Proteins are not attached but “caged” in
an aqueous environment. This approach has the advantage of displaying proteins in a more
“native” manner, but can support only a limited range of applications.
9809_C000.fm Page xvii Saturday, February 17, 2007 1:53 PM

by describing a variety of examples of applications enabled by fluorescent detection


technology. As useful as fluorescent detection has proven, however, there is a clear
need for “label-free” detection methods. This is especially true of small molecule
assays, where the addition of a fluorescent group can significantly alter the chemical
and biological properties of the compound under investigation. The most common
label-free detection technology, surface plasmon resonance (SPR) is described in
Chapter 10. This chapter reviews the basic physics behind the SPR phenomenon,
and discusses special considerations for the adaptation of SPR to arrays. Chapter 11
describes recent advances in the application of MALDI (matrix assisted laser des-
orption ionization) mass spectrometry to protein microarrays. In addition to detec-
tion, mass spectrometry can be used for molecular identification, potentially enabling
highly multiplexed experiments. Chapter 12 describes a recently commercialized
alternative to SPR based on photonic crystal biosensors.

APPLICATIONS
Functional protein microarrays have been adapted for a variety of applications in
both basic research and drug discovery. Two basic classes of experiments can be
performed with functional protein microarrays: interaction assays and activity
assays. Interaction assays profile the ability of molecules (or even cells) to bind to
proteins on the array surface. Activity assays profile the activity of proteins either
in solution or on the arrays themselves (see Figure 2). The breadth of applications
generated through these types of experiments is summarized in Table 1, and dis-
cussed in more detail in Chapters 13 to 19.
Chapter 13 describes the use of functional protein microarrays for profiling
protein-protein interactions, with a focus on 14-3-3 proteins. The use of protein

Interaction assays Activity assays


Molecular interactions Substrate assays Biochemical assays

B E
m m
A S E S

FIGURE 2 Basic types of assays. Interaction assays monitor the ability of a molecule/com-
plex (B) to bind proteins on the array (A). Activity assays monitor the activity of proteins in
solution, such as an enzyme (E) modifying (m) a substrate (S) on the array. Alternately, such
assays can monitor the activity of proteins on the array. (curved arrow represents a biochemical
reaction). (Reprinted with permission, Invitrogen)
9809_C000.fm Page xviii Saturday, February 17, 2007 1:53 PM

TABLE 1
Applications of Functional Protein Microarrays. A Summary of Many of the
Basic and Drug Research Applications of Functional Microarray Experiments
Experiment Basic Research Application Drug Research Application

Protein–protein interaction profiling Pathway mapping Target discovery


Protein–DNA interaction profiling Protein interaction mapping Early target validation
Protein–lipid interaction profiling Protein function
determination
Kd estimation

Substrate assays Pathway mapping Target discovery


Substrate identification Early target validation

Enzyme activity profiling Pathway mapping Target discovery


Enzyme activity discovery Early target validation

Protein–small molecule interaction Pathway mapping Target/mechanism


profiling Metabolomics determination
Chemical genomics Drug rescue
Alternate target identification
Specificity profiling
IC50 estimation
Lead optimization
Toxicity profiling

Antibody specificity profiling Antibody characterization Biotherapeutic development


and optimization

Immune response profiling Biomarker discovery Diagnostic


Biomarkers for efficacy and
safety
Vaccine design

Enzyme inhibitor profiling Enzyme characterization Specificity profiling


Lead selection and
optimization

Enzyme activity assay Enzyme kinetics Specificity profiling


IC50 determination
Lead selection and
optimization

Source: Adapted from Predki, P.F., Functional protein microarrays: ripe for discovery, Curr. Opin. Chem.
Biol., 8, 8, 2004. With permission.

microarrays for small molecule target identification is described in chapter 14, with
an emphasis on the author’s pioneering use of protein arrays to study chemical
genetics with the compound SMIR4. Chapter 15 discusses the possible uses of
functional protein microarrays for biotherapeutic drug development, with a particular
9809_C000.fm Page xix Saturday, February 17, 2007 1:53 PM

focus on using arrays to identify cross-reactive therapeutic antibodies, as well as


monitoring for autoimmune side effects. Chapter 16 describes the use of protein
arrays to discover antibody immune response biomarkers, an application which also
has implications for vaccine design and testing. The use of protein arrays to study
DNA binding is described in Chapter 17, including a discussion of the clinical
significance of such investigations. One of the most important and challenging
classes of proteins, multi-transmembrane spanning G protein-coupled receptors
(GPCRs), is addressed in Chapter 18. This chapter provides a thorough description
of this application, from surface chemistry to binding assay protocols and assay
validation. Finally, Chapter 19 describes the use of protein arrays for the identifica-
tion of kinase substrates, focusing on the application of this technique to the yeast
proteome.
While exhaustive coverage of all applications is not possible, this section
describes all of the major uses of protein microarrays currently under investigation.
No doubt, as the field evolves, new applications will be developed. The basics
described in this section, though, should provide a good foundation for understanding
these future developments.

DATA ANALYSIS
Data analysis is one of the most important, but often underappreciated, aspects of
the use of protein microarrays. This section starts with a thorough discussion of
image analysis in Chapter 20. Numerous considerations, from spot boundary assign-
ment and contaminant removal to statistical analysis and visualization, are described.
Chapter 21 takes off from there, describing approaches to analyzing the numerical
data generated directly from the images. Although focusing on biomarker discovery,
many of the approaches described in Chapter 21 are directly applicable to the other
applications described in this book. Chapter 22 uses computer simulations to help
evaluate the potential of protein microarrays for kinase substrate identification. Like
the previous chapter, however, the basic approach is applicable to many other
applications. The final chapter examines the software requirements for visualizing,
sharing and integrating the results of experimentation. It is only with this ability,
after all, that the full potential of this technology will be realized.

REFERENCES
1. Zhu, H. et al., Global analysis of protein activities using proteome chips, Science,
293, 2101, 2001.
2. MacBeath, G. and Schreiber, S.L., Printing proteins as microarrays for high-through-
put function determination, Science, 289, 1760, 2000.
3. Hall, D.A. et al., Regulation of gene expression by a metabolic enzyme, Science,
306, 482, 2004.
4. Ptacek, J., et al., Global analysis of protein phosphorylation in yeast, Nature, 438,
679, 2005.
5. Predki, P.F., Functional protein microarrays: ripe for discovery, Curr. Opin. Chem.
Biol., 8, 8, 2004.
9809_C000.fm Page xx Saturday, February 17, 2007 1:53 PM
9809_S001.fm Page 1 Monday, December 18, 2006 8:59 PM

Section 1
Functional Protein Content
for Microarrays
9809_S001.fm Page 2 Monday, December 18, 2006 8:59 PM
9809_C001.fm Page 3 Friday, January 12, 2007 3:01 PM

1 High-Throughput Gene
Cloning Using the
Gateway® Technology
Scott N. Peterson, Patrick Burr, Getahun Tsegaye,
and Pratap Venepally

CONTENTS

Introduction................................................................................................................3
Gateway Recombinational Cloning...........................................................................6
PCR Amplification of ORFs..........................................................................7
PCR Product Verification and Quantitation.........................................8
BP Clonase Reaction .....................................................................................9
E. coli Transformation...................................................................................9
Gateway Clone Resource Validation Procedure......................................................10
Sequence Assembly .....................................................................................13
Validation and Reports ................................................................................15
Destination Vector Cloning .....................................................................................16
Technology and Robotics ........................................................................................17
Methods and Materials ............................................................................................19
PCR Amplification of ORFs........................................................................19
PCR Product Verification and Quantitation ................................................19
BP Clonase Reactions .................................................................................20
DH10B-T1 E. coli Transformation .............................................................20
Clone Sequence Validation..........................................................................20
Plasmid Extraction.......................................................................................20
Sequencing Template Production by TempliPhi .........................................21
References................................................................................................................21

INTRODUCTION
The genomic era has produced an ever-increasing number of complete genome
sequences from a wide variety of organisms. The large number of annotated gene
sequences being produced has driven the advancement of numerous complementary
technologies that enable research scientists to exploit the availability of genome

3
9809_C001.fm Page 4 Friday, January 12, 2007 3:01 PM

4 Functional Protein Microarrays in Drug Discovery

sequence data in new and powerful ways.1 One such technology is the Gateway®
cloning system made available by Invitrogen Inc.2
The introduction of this technology was particularly well timed in relation to
genomic sequencing, since the Gateway platform provided a vehicle for the cloning
and expression of complete open reading frames (ORFs). Prior to the introduction
of the Gateway cloning technology, classical cloning strategies using restriction
enzymes and DNA ligase were fully entrenched. The primary limitation of traditional
cloning procedures was the difficulty in implementing high-throughput approaches.
The displacement of the traditional methods would require a clear and substantial
improvement in efficiency, ease of use and automation potential. The Gateway
technology delivered these requirements. The increased use of this cloning system
is in turn driving the development of a novel series of technologies that these
expression clones feed directly. Foremost are those technologies associated with in
vivo and in vitro protein expression and purification for functional and structural
analysis of proteins. The improved efficiency and ease of generating the raw mate-
rials (DNA expression clones, purified recombinant proteins) are supporting a vig-
orous growth in the use of immobilized proteins on glass surfaces. These advances
hold promise for accelerating the discovery of functional roles of genes and provide
new strategies for identifying drug targets and therapeutics.
In response to the challenge put forward by the Nation Institute for Allergy and
Infectious Disease (NIAID) to generate and distribute cloned ORFs to the scientific
community, to enable functional genomics of microbial pathogens, viruses and
parasites, the Pathogen Functional Genomics Resource Center (PFGRC) at The
Institute for Genomic Research (TIGR) was motivated to identify a cost-effective
and efficient cloning technology. It was important to select a strategy that not only
provided the necessary efficiency for high-throughput cloning, but also one that
would be widely recognized and well-accepted by the diverse scientific end-user.
The widespread adoption of the Gateway cloning platform was fortuitous since major
cloning efforts performed in other laboratories are now commonly using the Gateway
platform and have therefore enabled the collaboration and clone sharing among
scientists with diverse scientific interests (see appendix for other users of the tech-
nology). While individual applications may vary, the primary purpose of the Gateway
platform is the generation of cloned ORFs in one or more expression vectors
(destination vector). This is accomplished in two steps that together mimic the
recombination reaction that occurs between the genome of E. coli and that of phage
lambda. The lambda phage genome contains an attP site that undergoes recombi-
nation, with the aid of lambda phage and E. coli–encoded proteins with the 25 bp
attB site in the E. coli genome. Upon recombination, the attP site divides in two
halves, attL and attR, that flank the lambda genome. During lytic phase, lambda
undergoes a second recombination between attL and attR thus reconstituting the
attP site while leaving behind the original attB site. In the context of gene cloning,
a PCR product (ORF) is generated that is flanked by two nonidentical, primer
encoded attB sites (attB1 and attB2). The PCR product undergoes recombination
with a vector containing two nonidentical attP sites (attP1 and attP2). The recom-
bination reaction is efficient and directional. The clones derived from this recombi-
nation reaction are referred to as entry clones. Entry clones contain inserts that are
9809_C001.fm Page 5 Friday, January 12, 2007 3:01 PM

High-Throughput Gene Cloning Using the Gateway® Technology 5

attB1 ORF attB2

Recombination
ORF
event attB1 attB2
ccdB cmr Viral ORF
attP1 attP2
attB1 attB2
pDONR 221 His tag

ccdB ampr
Entry clone attR1 attR2
attL1 attL2 Destination vector
ORF
Example
Promoter
ccdB ampr GFP ORF
attR1 attR2 attB1 attB2

GFP destination vector Your expression


ORF vector
attB1 attB2
2 hybrid
Promoter
ORF GFP
attB1 attB2
GFP expression vector

FIGURE 1.1 Gateway Cloning by Recombination.

flanked by two nonidentical attL sites. The entry clone has no direct function other
than to serve as the substrate for the transfer of the ORF into an expression vector
via an LR reaction, named because the attL sites in the entry clone recombine
directionally with attR sites in the destination vector. The entry clone is considered
to be a useful resource since the cloned insert can be readily shuttled into any number
of commercially available or user-designed expression vectors (destination vectors)
without a priori knowledge of the intended down-stream application (Figure 1.1).
The generation of entry clones in a high-throughput manner is a multistep
process that, taken together, results in a high overall cloning efficiency. This effi-
ciency can be attributed to two features. First, recombination of att sites in both BP
and LR reactions is nearly stoichiometric and requires the input of a purified PCR
product and vector DNA into a proprietary mixture of enzymes that catalyze the
recombination of the PCR product into the cloning vector. Second is the system’s
use of both negative and positive selection in the subsequent transformation of E. coli.
The successful recombination between vector and PCR product displaces the resident
“stuffer fragment” that consists of the markers ccdB and Cmr. The ccdB gene product
interferes with gyrA activity and is therefore toxic to E. coli. Nonrecombinant vector
will retain the ccdB and therefore not be frequently recovered following E. coli
transformation. The vector backbones used contain standard antibiotic resistance
genes for positive selection of transformants. The vast majority (>99%) of colonies
that form are recombinant clones. The simplicity of the cloning reaction allows full
automation of the steps leading up to and including the cloning reaction itself.
9809_C001.fm Page 6 Friday, January 12, 2007 3:01 PM

6 Functional Protein Microarrays in Drug Discovery

Primer design

PCR optimization PCR setup

PCR Failed reactions

Product validation

BP reaction

E. coli transformation Electroporation

2 colonies

Sequencing template prep

Glycerol stocks End reads

Collapse to validated set Walking primers

Sequence validation

FIGURE 1.2 High-Throughput Clone Production Pipeline.

We have developed a nearly fully automated pipeline for the cloning and sequence
validation of ORFs using Gateway. The automation not only provides the potential
to generate large numbers of recombinant clones but also the implementation of a
process for the tracking of materials through a Laboratory Information Management
System (LIMS). The development of a functional LIMS serves to reduce sources of
human error and reagent waste. The ability to automate the process is very important
to our pipeline since, for several steps, second and even third attempts are made on
a small number of failed cases requiring “cherry-picking” and subsequent reintegra-
tion with the complete clone set. The creation of a fully functional and automated
clone validation sequence analysis process has led to increased throughput and
efficiency of clone production. The Gateway clone production pipeline (Figure 1.2)
illustrates the integration of this multistep process.

GATEWAY RECOMBINATIONAL CLONING


Primer Design: Each of the unique open reading frames (ORFs) identified and
annotated in a genome are potential targets for forward and reverse primer design.
Recently duplicated genes displaying a high degree of sequence identity may be
difficult, if not impossible, to amplify in pure form. Each forward primer contains
a 5′, 25 nt attB1 sequence (see Materials and Methods) appended to each gene
specific sequence representing the start codon and the neighboring 3′ nucleotides
required to achieve a predefined Tm = 60–65°C. The reverse primer has a 5′, attB2
sequence appended to gene specific sequence beginning at the nucleotide just
upstream of the stop codon. This design feature allows the subsequent flexibility to
9809_C001.fm Page 7 Friday, January 12, 2007 3:01 PM

High-Throughput Gene Cloning Using the Gateway® Technology 7

create COOH-terminal fusion proteins, wherein a stop codon is conferred by the


cloning vector, just downstream of the cloned ORF in each of the three possible
reading frames. Some investigators prefer to include a stop codon in the PCR primer.
This is accommodated by altering the primer design to include either an endogenous
stop codon or a standard (uniform) stop codon. Each primer contains four G residues
at the 5′ end. These residues are important for recombination efficiency and serve
to internalize the attB sequences so that they do appear at the very end of PCR
products. We sort our primer pairs with respect to the anticipated PCR product size
from smallest to largest. By restricting the Tm of each primer used within a small
range, we can define efficient cycling conditions based on the single variable of
extension time. For whole genome applications the range of size of ORFs arranged
by size in any 384 grouping is relatively small allowing us to define extension times
that are nearly optimal for all targets. We have had very good success using oligo-
nucleotides obtained from Illumina Inc. The forward and reverse primers are syn-
thesized in identical well locations in paired 96-well plates, facilitating manual or
automated robotic setup of PCR reactions.

PCR AMPLIFICATION OF ORFS


The production pipeline developed in the PFGRC is quite generalized and its overall
efficiency is not strongly influenced by the specific ORFs to be cloned. One exception
is the prior optimization of PCR conditions for the genomic DNAs of interest. The most
pronounced variable to account for is the G+C content of the genome. We have found
that species by species optimization of the strategy used for amplification of ORFs using
the PCR is critical, especially when large numbers of reactions are to be performed. We
have identified four proof-reading polymerases that when applied to particular genomes,
perform well (Table 1.1). This list is by no means exhaustive but provides a guideline
for robust polymerases for use in a high-throughput cloning process.
Once PCR reaction optimization is complete, high-throughput reaction setup
and cycling is ready to begin. We perform PCR in a 35 µl reaction volume in 384-
well format. The scale of the reaction provides sufficient yield of product for
subsequent cloning reactions. Primer dimers containing both attB sites represent
clonable products and therefore behave as active competitors with the ORF in BP
cloning reactions. An alternative process that we have not investigated thoroughly
is the use of a two-step PCR reaction. The primers used differ from those described
above and include only the 3′ half of the attB sequence. After a limited number of
PCR cycles, the products are cleaned-up to remove the initial primers and a second
set of universal primers containing a complete attB site are used in all reactions.
Since the second primer pair is used in all reactions, the cost of primer synthesis
can be driven down. After cycling, the PCR products are transferred to 384-well
filtration plates (Millipore) using a Beckman Coulter Biomek-FX 96 probe liquid
handling robot. For lower throughput applications, a multichannel pipette is a useful
alternative. PCR products are purified according to the manufacturer’s suggested
procedure and products are eluted in 50 µl of H2O and finally transferred to a clean,
384-well MJ Research hardshell plate using a Beckman Coulter Biomek-FX 96
probe liquid handling robot.
9809_C001.fm Page 8 Friday, January 12, 2007 3:01 PM

8 Functional Protein Microarrays in Drug Discovery

TABLE 1.1
PCR Kits for High-Throughput ORF Amplification
Product Manufacturer Catalog Description Target Genome

Phusion High- Finnzymes New F-530-L Proprietary Pyrococcus- H. pylori,


Fidelity DNA England like enzyme with B. anthracis,
Polymerase Biolobs processivity enhancing S. agalactiae,
domain and S. typhimurium,
proofreading capacity S. pneumoniae,
V. cholerae,
Y. pestis
error rate reported:
(4.4 × 10–7)
Platinum PCR Invitrogen 12532−016 Complex of recombinant S. aureus COL
SuperMix Taq polymerase and
High Fidelity Pyrococcus species
GB-D with proofreading
capacity
error rate reported:
6 × less than native Taq
(approx. 1.0 × 10–5)
Takara LA Takara RR013 Proprietary modified Taq M. tuberculosis
PCR kit polymerase with
proofreading capacity
error rate reported:
6.5 × less than native Taq
(approx. 1.0 × 10–5)
Advantage –HF BD Biosciences K1909-1 Proprietary mix of a M. tuberculosis
PCR kit modified Taq
polymerase with a
Pfu-like proofreading
polymerase
error rate reported:
20 × less than native Taq
(approx. 5.0 × 10–6)

PCR Product Verification and Quantitation

The purified PCR product yield is determined using a Caliper ASM90 SE capillary
electrophoresis instrument (Caliper LifeSciences). The Caliper System uses a “sipper”
mounted on a robotic arm to remove ~1 µl from each well. Each PCR product is
electrophoresed through the single capillary, where its mobility is compared to a set
of size standards. The quantity and relative purity (single band) of each PCR frag-
ment is determined in a matter of 30 seconds. The size estimates in our experience
are accurate to ±5%. PCR products deviating by more than 10% from an expected
size are flagged. In less than 1% of the cases we observe size estimates outside this
range. Interestingly, a significant proportion of these are ultimately determined to
9809_C001.fm Page 9 Friday, January 12, 2007 3:01 PM

High-Throughput Gene Cloning Using the Gateway® Technology 9

TABLE 1.2
Summary of ORFeome Projects
Validation Validation
PCR Transformation Success Success Overall
Project 1st Round Follow-up Heat Shock Electroporation Colony 1 Colony 2 Success

S. aureus 93.1% 94.4% 98.0% 99.9% 74.3% 14.3% 88.6%


COL
F. tularensis 87.3% 94.6% 96.4% 99.3% 65.8% 18.8% 84.6%
SHU S4

be the expected ORF with no structural rearrangements. The DNA yield, size, and
purity are stored directly in the instrument’s online computer where it is then
classified as passing or failing. Common reasons for scoring a reaction as a failure
include low or no yield (<10 ng/µl) or the formation of two or more PCR products
(poor primer specificity). Failed reactions (~10%) are identified automatically in a
report form that is used to “cherry-pick” appropriate primer pairs for a second pass
attempt. A second attempt to amplify failed PCR reactions, using either the identical
reaction conditions or those of another kit, generally results in an additional 1 to
8% increase in overall success achieved (~95%). In our experience 0.5 to 1% of
PCR failure is attributable to oligonucleotide synthesis. Resynthesis of oligonucleo-
tides for failed reactions imposes additional economic burden on a project with
limited returns. A summary of two recent whole genome microbial ORF cloning
projects are shown in (Table 1.2). Successful PCR products are then merged together
before proceeding to the BP reaction.

BP CLONASE REACTION
The output file generated by the Caliper contains the concentration of each PCR
product that is then converted to a molar concentration. This information is fed
directly into the Biomek-FX SPAN-8 liquid handling robot for automated set up of
BP cloning reactions. We have adopted the use of a scaled down version of the BP
reaction using 50 fmol of target vector and PCR product insert. A master mix
containing all components other than the PCR product insert is prepared and ali-
quoted into individual wells. Each PCR product is diluted to a concentration of
25 fmol/µl, and subsequently 2 µl of each are added to the master mix. The BP
cloning reaction efficiency is inversely proportional to the size of the PCR product
to be cloned. This relationship is only strongly limiting for very large genes (~5 Kb).
In an earlier version of our pipeline we set up BP reactions at 2 separate scales 25
fmol for PCR products <2.5 Kb and 100 fmol for PCR products >2.5 Kb. More
recently we have used 50 fmol scale reactions for all PCR products.

E. COLI TRANSFORMATION
We have not yet identified a reliable 96-well device for electroporation of electro-
competent E. coli cells. The efficiency of most cloning efficiency chemically
prepared competent cells is more than adequate for recovery of recombinant clones.
9809_C001.fm Page 10 Friday, January 12, 2007 3:01 PM

10 Functional Protein Microarrays in Drug Discovery

E. coli transformations are conducted in 96-well trays and are set up robotically.
After heat shock and recovery in nonselective media for 1 hour, cells are robotically
plated onto 20 × 20 cm dishes that are sub-divided into 48 grids. Each grid is
preseeded with 6 to 8 glass beads (3 mm) and when all 48 transformation reactions
are distributed, the plate is shaken gently until all liquid is absorbed into the solid
media. The glass beads are removed by inverting the plate into an appropriate
receptacle. Any grid yielding one or fewer colonies is considered a failure. A list of
failed transformations is compiled and those BP reactions are used a second time
to manually transform electro-competent E. coli cells. An interesting feature we
observe is that virtually all of the PCR reactions scored as failures lead to colony
generation, following this two step procedure. Slightly more than half of these cases
result in valid full-length recombinant clones.

GATEWAY CLONE RESOURCE VALIDATION


PROCEDURE
The validation of Gateway clones is an important aspect of clone production. It is also
the most challenging to conduct. Since Gateway clones are most commonly used for
expression, it is important to validate the clones for sequence and length integrity. We
have not observed substantial DNA rearrangements of cloned inserts; however,
nucleotide substitutions introduced during PCR are frequent enough that further con-
sideration is warranted. Thus far, no standards exist for clone validation. In an attempt
to initiate such a standard we have adopted a standard set by The Harvard Institute
for Proteomics (HIP). The HIP sequence validation standard is stringent but reasonable
and involves rejecting any clones containing more than 2 nonsilent substitutions.
Clones containing indels (frame shift), nonsense, and/or mutations in the att sites are
rejected. These validation criteria have driven our validation strategy to include a
two-tier process that begins with the sequence validation of a single colony. If that
DNA insert fails the validation criteria, a second colony is then analyzed. Mutations
introduced early in PCR cycling will be present in a large fraction of the resulting
colonies; however, in practice we find a relatively high degree of utility in going to a
second colony in instances where the first colony was deemed unacceptable. Data in
Table 1.2 illustrate the utility derived from the analysis of a second colony and our
future interest to determine the point of diminished returns in terms of the number of
colonies to select for sequence validation. This option must be weighed against the
alternative that is to begin the process again from the start.
Initial attempts to sequence validate Gateway entry clones resulted in unacceptably
low sequencing success frequencies. This was particularly evident for clones
containing small inserts, <600 bp, but also negatively affected end reads obtained
for larger inserts. It was suggested that the attL sites flanking the cloned inserts
have significant potential to form secondary structure that polymerases have dif-
ficulty traversing. Specific blocking primers were designed to inhibit the secondary
structure formation, thus partially alleviating the barrier to polymerase processivity.3
We have verified the utility of the blocking primers and observed discrete
improvement to our overall entry clone sequencing efficiency. Despite the
improvement afforded by the use of blocking primers, our sequencing success was
9809_C001.fm Page 11 Friday, January 12, 2007 3:01 PM

High-Throughput Gene Cloning Using the Gateway® Technology 11

still below that routinely obtained for other cloning vectors (~90%). We have
developed further improvements to the sequence validation of entry clones using
phi29 polymerase (Amersham, Inc.) on crude lysates prepared directly from colonies.
Templates prepared in this manner are advantageous, although for reasons that are
not completely understood. The use of sequencing templates generated through
the random priming of plasmid DNAs by rolling circle amplification alleviate the
observed sequencing failure for clones containing inserts <600 bp. This strategy
has allowed us to generate sequencing results consistent with expected success
frequencies and quality. For cloned inserts 600 bp and larger we have had com-
parable high frequency sequencing success using either templates derived from
templiPhi or double-stranded plasmid DNAs. A remaining dilemma in sequence
validation of clones involves confirmation of the correctness of the attB site itself.
Sequence traces from failed validation attempts are often due to abrupt termination
in signal strength as the polymerase reaches the att site. Yet another possible
solution to sequence validation is to direct efforts on the destination clone which
is flanked only by attB sites.
The generation of DNA templates for sequence validation using templiphi is
simple, inexpensive, and easily automated. A sample of 10 µl from an overnight
culture is used to prepare lysates by brief heat treatment, 93°C for 3 minutes. A small
volume from the cleared lysate is used as template for templiphi reactions. The
reaction products are diluted to a final volume of 40 µl with H2O. We typically
obtain yields between 20 and 40 ng/µl, which is sufficient for approximately 20
sequencing reactions.
The number of sequencing reactions performed to validate a cloned ORF is
based on its length. For ORFs 500 bp or less, only end-reads are performed. For
ORFs larger than 500 bp internal walking primers are designed to generate reads in
both directions. The optimal density of walking primers is dependent on average
read length and overall sequencing success. We have compared the outcomes of
applying walking primers at a regular spacing of 250 bp and 500 bp (Figure 1.3).
Given that an average read length from automated sequencing instruments is now
in excess of 800 bp, it may be surprising that walking primers at such high density
are required for high-throughput sequencing. The failure of some sequencing primers
and reactions is a given and if the spacing of sequencing primers is too great, these
failures will result in an inability of neighboring primers to fill the gap. The first-
pass sequencing attempt (end reads and walking primers) results in a number of
outcomes ranging from perfectly validated clones (2X coverage, no mutations) to
assemblies with only partial coverage. The classification of sequence validated clones
is described below (Table 1.3). We can see that 500 bp spacing among walking
primers compares unfavorably to 250 bp spacing in terms of the frequency at which
validated clones are identified. From our perspective the choice between 250 bp and
500 bp walking primers spacing is a matter of decision drivers like economics and
throughput. The average gene requires seven walking primers for validation and
therefore represents a substantial cost. Reducing these costs by nearly 50% is
potentially attractive but does carry the consequence of increasing the amount of
second-pass sequencing attempts required to fully validate a clone. These additional
attempts also carry a cost.
9809_C001.fm Page 12 Friday, January 12, 2007 3:01 PM

12 Functional Protein Microarrays in Drug Discovery

a. 250 pp ORF length # of


primer pairs
A 5' 3'
250 250 250 250 250 250 1–249 0
b. 500 pp 250–499 1
5' 3' 500–749 2
250 250 250 250 250 250
c. 250-f_only 750–999 3
5' 3' 1000–1249 4
250 250 250 250 250 250
1250–1499 5
d. 250-r_only ......
5' 3'
250 250 250 250 250 250
B
45
A+B A+B+C
39 A
40 250 pp 100 100 B
500 pp 82.97872 96.9697 C
35 250-f_only 89.3617 98.48485 D
Number of clones

250-r_only 91.48936 98.48485


30
27
25 25
25 23 23 22
19 19
20
16
14
15
10 8
5
2 1 1
0
0
250 pp 500 pp 250-f_only 250-r_only
Primer interval
AV_read_Ingth: 940 838 939 942 842 936 917 941 853 941 857 944 860 944 98

FIGURE 1.3 The effect of walking primer spacing on sequence validation. (A). Primer design
schema (left). Walking primer pairs (forward and reverse arrows) at an interval of 250 (a) or
500 bases (b) or single alternating forward and reverse primers at 250 base intervals (c,
forward first primer; d, reverse first primer) are used for sequencing. The circles (•) demarcate
250 base intervals. The table on the right lists number of primer pairs designed for various
ORF length intervals. (B) Primer intervals vs. sequence coverage. Number of clones in each
single-contig sequence validation class A, B, C, and D (see text and Table 1.3) obtained with
different walking primer intervals, illustrated in Figure 3A, are plotted. The inset table shows
the percentage of full-length A and B wild-type class clones and A, B and C class (full-length
clones with mutations) clones seen with different walking primer intervals relative to 250
base interval primer pairs (set to 100). The average length of sequences considered in the
validation of clones in each class is shown at the bottom.
Depending on the nature and number of remaining clones, directed efforts are
applied to close remaining gaps and confirm sequence ambiguities in the assembled
sequence. After applying brute force and manual efforts to elusive clones, a final
sequence validation report is generated that directs the representation of acceptable
clones for distribution to the scientific community. The list of acceptable ORFs is
used to direct the robotic compression of the two freezer copies (colony 1 and 2),
into a final set that is replicated into several glycerol stock copies.
9809_C001.fm Page 13 Friday, January 12, 2007 3:01 PM

High-Throughput Gene Cloning Using the Gateway® Technology 13

TABLE 1.3
Sequence Validation Classes
Valid Classes: The clone has either a Invalid Classes: The clone has either a
full-length or partial-length full-length or partial-length
coverage and shares greater coverage and shares less than
than 90% sequence identity 90% sequence identity with
with the wild-type reference the wild-type reference

A — Full-length sequence — 2x or greater M — Full or partial-length sequence — less than


sequence coverage at each base, 100% sequence 90% sequence identity with the reference ORF.
identity with the reference
B — Full-length sequence — 1x or greater N — Full or partial-length sequence — less than
sequence coverage at each base, 100% sequence 100% sequence identity with the NON-reference
identity with the reference ORF
C — Full-length sequence — sequence variation T — No good quality sequences available for
(< 100% but > 90% sequence identity with the sequence assembly and validation
reference)
D — Partial-length sequence — single contig U — Sequencing status unknown
with missing end-sequence (> 90% sequence
identity with the reference)
E — Partial-length sequence — multiple contigs W — Full-length sequence — 1 × or greater
with gaps in assembly (> 90% sequence identity sequence coverage, 100% sequence identity
with the reference) with the NON-reference ORF
Z — Failed assemblies due to process errors in
the assembly pipeline

Note: Clones are grouped into various valid and invalid classes based on the identity and the coverage
of the cloned sequence vis-à-vis the reference.

SEQUENCE ASSEMBLY
The sequences obtained from Gateway entry clones are validated using a novel high-
throughput assembly pipeline, called CLASP (CLone validation ASsembly Pipeline)
developed by the PFGRC bioinformatics group at TIGR. This software can be
accessed upon request (www.pfgrc.tigr.org). Initially, the algorithm performs the
assembly of individual sequences, generated from the cloned insert, to form a
consensus sequence. Subsequently, the consensus is compared to the reference ORF
and a validation report detailing the quality of the cloned sequence is generated.
The clone assembly validation pipeline exploits the fact that the sequence of
the insert in the vector is already known. As shown in the Figure 1.4, it uses two
separate assembler programs to optimize the assembly of the sequence reads and
achieve maximum accuracy in the consensus sequence. The first and the most
important of these is AMOScmp (Figure 1.4A) which places a premium on the
agreement between the reads and the reference — rather than on the phred quality
scores generated from the trace files. This guides the selection of which sequences
to use for the final contig and thus the consensus. The AMOScmp assembly allows
joining of even short overlapping sequences, resulting in a high recovery of single,
9809_C001.fm Page 14 Friday, January 12, 2007 3:01 PM

14 Functional Protein Microarrays in Drug Discovery

Assembly
(A) AMOScmp (B) Minimus

Collect aligned & not-so-well aligned


Collect aligned reads, adjust clear ranges reads, adjust clear ranges to maximum
to the end of alignment in both directions (aligned or default ‘clear’ values,
whichever is greater, in both directions)

Assemble Assemble
2 or more contigs Yes
1 contig? E
1 contig? M/N/W
No No
Yes
Contigs Yes
No Single contig with gap(s) overlap?
A, B, or C? D
at the ends No
Yes
AutoJoin*
AutoEdit*
Contigs Yes
overlap?
Compare minimus
AddCoverage*
& AMOScmp results to join contigs
No
Final output
AutoEdit*

FIGURE 1.4 A schematic diagram of the CLASP assembly pipeline. Two assemblers — one
comparative (AMOScmp) and another noncomparative (Minimus) — are used serially to assemble
sequencing reads from the inserts cloned into the Gateway vector. The blue circles indicate the
ends of individual sequences defined as ‘clear’ (good quality). AMOScmp does not rely on these
but utilizes the alignment with the reference sequence to determine the extent of reads for the
assembly of contigs (left panel, A). Minimus (right panel, B) is used to assemble reads that do
not align well with the reference sequence. See text for the definitions of various validation classes.

full-length contigs (classes A, B and C, Table 1.3) which otherwise might remain
as a single (class D) or multiple (class E) partial-length contigs. For sequences that
align very poorly with the reference sequence and result in either partial-length
contigs or no contigs after applying AMOScmp, a second assembler, called Minimus
(Figure 1.4B), is used. In Minimus the amount of sequence used from any read in
the assembly are determined by the phred quality scores instead of how well they
align with the reference. For that reason, the consensus sequence(s) obtained from
the Minimus assembly have less identity with respect to the reference sequence than
those generated by the AMOScmp assembler (classes M, and N; Table 1.3). However,
the Minimus results are indispensable for (a) ‘catching’ clones which are good but
mislabeled in any of the steps along the cloning process (class W) and also in the
(b) identification of clones that do not meet the acceptance criteria for valid clones
and thus require repeat efforts at cloning and validation. After the assembly with
AMOScmp or Minimus, the base calls in the consensus sequence(s) obtained for
each clone is verified for accuracy against chromatograms using autoEditor*. In the
case of multiple contigs with gaps, autoJoiner* is used to join the neighboring reads

* Sequencing closure software developed by TIGR.


9809_C001.fm Page 15 Friday, January 12, 2007 3:01 PM

High-Throughput Gene Cloning Using the Gateway® Technology 15

651 2897
650 TTG TAC AAA AAA GCA GGC TNN NAC CCA GCT TTC TTG TAC AAA
Gene
AAC ATG TTT TTT CGT CCG ANN NTG GGT CGA AAG AAC ATG TTT
attL attR

FIGURE 1.5 The recombination sites in the Gateway entry vector. The sequences of attL
(attB1) and attR (attB2) sites, verified for their integrity along with the enclosed insert, are
indicated by the underlines.

by relaxing (extending) the clear ranges if they align well with each other above a
set threshold value (Figure 1.4B). The final contig(s) for the clones, which are
processed by both assemblers, is chosen based on the best coverage in length and
the identity shown vis-à-vis the reference.

VALIDATION AND REPORTS


The consensus sequence generated in the assembly process is analyzed not only
for the integrity of the insert but also of the flanking attL sites up to BsrGI sites
(5’ TGTACA 3’ sites at 651 on the forward strand and at 2903 on the reverse strand —
Figure 1.5.) Following sequence validation by BLASTN and BLASTX analysis
against reference nucleotide and protein sequences, respectively, the clones are
classified into valid and invalid categories as defined in Table 1.3. The details of the
final validation data are presented in two reports. One of them, clone_distribution_
report.html (Figure 1.6, partly shown), shows the details of validation for each clone
including the sequences for the cloned insert and the reference ORF — via hyperlinks
shown in the last column. The Class and Mutations fields in each case are hyperlinked

FIGURE 1.6 A partial screenshot of clone_distribution_report.html. Various aspects of the


validation data for the clones are displayed in the file. Sequences for the cloned insert and
the reference sequences (shown at the bottom) are accessed via hyperlinks present in the last
column of the file. See text for the details.
9809_C001.fm Page 16 Friday, January 12, 2007 3:01 PM

16 Functional Protein Microarrays in Drug Discovery

FIGURE 1.7 Clone alignment and summary of mutations. (A) Nucleotide and protein alignments
(top and bottom panels) and a summary of mutation(s) (middle panel) are shown. See text
for the details. (B) A partial screenshot of clone_distribution_report_C_class_ mutations.html.
Various categories of C class clones, grouped based on the number and types of mutations
at the protein level, are shown (links at the top and the details of a link at the bottom are
shown as an example). In addition, clones categorized on the basis of whether the mutations
occur in the CDS or the flanking “att” sites, or both at the nucleotide level are shown in the
middle (only links shown).

to the files showing nucleotide and protein alignments with the reference sequence
(Figure 1.7A, top and bottom panels, respectively) and the summary of mutations,
if any (Figure 1.7A, middle panel). Positions of mutations and the associated con-
sensus base call quality values, calculated using the procedure described by Churchill
and Waterman,4 are displayed below each nucleotide alignment. In addition, a quality
class is assigned to each mutation suggesting the level of confidence in the base call
(Figure 1.7B, middle panel). A second report (clone_distribution_report_C_ class_
mutatations.html) shows the further sub-division of C class clones based on whether
the mutations occur within the coding DNA sequence (CDS) or the flanking ‘att’
sequences and whether they represent silent or missense or nonsense mutations at
the protein level.
The first pass attempt results in a number of outcomes ranging from perfectly
validated clones (classes A and B, see above) to assemblies with only partial cov-
erage. In cases where the first pass validation attempts indicate that a clone has more
than 2 nucleotide substitutions relative to the reference sequence, or no cloned insert,
the second colony, held as a glycerol, stock is used for template production and
sequence validation. Upon generating comparable sequence data from the second
colony and the validation process, additional clones which pass the acceptable
9809_C001.fm Page 17 Friday, January 12, 2007 3:01 PM

High-Throughput Gene Cloning Using the Gateway® Technology 17

criteria are identified and consolidated with those selected from the first colony. At
this juncture, depending on the nature and number of remaining invalid clones,
further sequencing on both first and the second colony templates are performed in
an effort to close remaining gaps and resolve any sequence ambiguities in the
assembled sequence.
Following the final consolidation of the validated clone set, reports with various
details on the validation are generated. The list of acceptable ORFs from these reports
is used to direct the robotic compression of the two freezer copies (colony 1 and 2),
into a final set that is replicated into several glycerol stock copies (5 to 10). Since
accurate storage and retrieval of samples is essential to a facility managing the
distribution of thousands of clones, PFGRC has acquired and installed a Biophile
Storage and Individual Vial Retrieval System (Biophile, TekCel) for this purpose.
The system consists of five –80°C storage units (BSU) and a –40°C individual vial
retrieval unit (IVR). The system is integrated into a relational database, utilizing bar
code information to identify each clone and its location in storage. The IVR system
automatically records, stores, and retrieves each requested vial based on a prepared
worksheet. The “rearraying” of individual vials into new sets allows accurate retrieval
of clones or clone sets.

DESTINATION VECTOR CLONING


The ability to automate a large portion of the Gateway Clone Resource production
pipeline accounts for the high-throughput capabilities that the PFGRC now offers.
The ease of use of the Gateway technology makes possible the construction and
validation of 10,000 or more expression clones annually. The scale up of the
procedure is largely dependent on the acquisition of additional robots. The ability
to transfer cloned and validated entry clone inserts into Gateway expression
vectors is straightforward. Purified entry clones and destination vectors are mixed
and recombination occurs faithfully via an LR clonase reaction. The screening
of recombinant clones and validation of their quality can be performed in a
streamlined manner as well, since there is no need to repeat the sequence vali-
dation, the only important check being to establish that the complete ORF is
present in the expression vector. In general, a single colony can be selected for
insert validation. Direct PCR from selected colonies using forward and reverse
Gateway® primers allow a rapid and cost-effective method for qualifying the
expression clones.

TECHNOLOGY AND ROBOTICS


The PFGRC utilizes two versions of the Beckman, BioMek FX platform to automate
most steps in the process, including PCR reaction setup, PCR product purification,
cloning reaction setup, transformation of chemically competent cells, plating of trans-
formed cells, plasmid isolation, setup of sequencing plate format, and replication of
clone stock copies for distribution. The FX-96 platform transfers equal volumes of
9809_C001.fm Page 18 Friday, January 12, 2007 3:01 PM

18 Functional Protein Microarrays in Drug Discovery

96 samples in parallel, and is used for processes that have pre-equalized concentra-
tions of all reagents, such as PCR reaction setup, and clone stock replication. The
FX-Span8 platform transfers samples individually with one of eight pipetting tips
with independent volume and sample well location control. Additionally the range
of movement and software flexibility allows the deposition of transformation reac-
tions onto oversized 48-well, sectored agar culture plates for colony isolation.
Together these two instruments have provided the necessary flexibility and accuracy
to make high-throughput Gateway cloning efficient and reliable. One adaptation to
reaction setup necessitated by the use of robotics is to ensure that each individual
pipetting step delivers 2 µl or more, since overall pipetting accuracy below these
volumes is lower.
Accurate qualitative and quantitative assessment of PCR amplification products
is an essential part of the high-throughput application of Gateway technology. PCR
products must be screened for the presence of multiple bands, incorrect size products,
and failed reactions. The precise size (bp) and concentration of successfully ampli-
fied products must be known to ensure that optimal recombinational cloning effi-
ciency is achieved. The PFGRC utilizes the Caliper ASM 90 SE Capillary Electro-
phoresis platform for this purpose. Other similar technologies offered by Agilent
Inc and others, perform in a similar way. The Caliper instrument has the ability to
accomplish these tasks in parallel with a high degree of accuracy and walk away
automation. The system performs electrophoresis in a single gel filled micro capillary
channel with high resolution and processing speed (100 to 5000 bp, 30 seconds per
sample), allowing the characterization of a 384-well microtiter plate in approxi-
mately three hours. The PCR products are detected using a fluorescent dye that
provides high sensitivity detection of secondary products and smears that could go
unnoticed using traditional agarose or polyacrylamide slab gels. Sizing of the
detected PCR products is automated and accurate within ±5%. The concentration
of any bands detected is also calculated automatically based on a standardization
sample. The area under the peak is calculated for product bands and compared to
the standard. This method provides results that are more accurate than traditional
absorbance readings taken at 260 nm as it relies on an intercalating fluorescent dye
rather than absorbance that can be skewed by multiple factors. The output from the
system is then transferred to liquid handling robots for subsequent automated reac-
tion set up.
To achieve a high level of success in a high-throughput endeavor such as the
Gateway clone validation pipeline, tracking various laboratory and data processing
steps in a systematic way is very critical. A software system like LIMS (Laboratory
Information Management System) or a similar resource will be very helpful in that
effort and can aid in the creation of high-quality Gateway clones in the following
ways: (a) by capturing measurement data, a LIMS can ensure that the correct values
are used for PCR evaluation and calculations, (b) by automatically generating robot
rearray scripts, a LIMS can prevent plate-to-plate transfer errors, as well as speed
up lab processing, and (c) by providing analysis and reporting tools, it can provide
valuable metrics that allow lab personnel to evaluate the quality of their techniques
over time to improve them.
9809_C001.fm Page 19 Friday, January 12, 2007 3:01 PM

High-Throughput Gene Cloning Using the Gateway® Technology 19

METHODS AND MATERIALS


PCR AMPLIFICATION OF ORFS
Forward and reverse primers are designed to amplify each ORF from a reference
genome sequence. Each oligonucleotide sequence is then appended 3’ of the attB1
(forward) and attB2 (reverse) sequence.

Forward Primer:
5′ GGGG ACAAGTTTGTACAAAAAAGCAGGCTTC (N18-25) Gene Specific Seq 3’
Reverse Primer:
5′ GGGG ACCACTTTGTACAAGAAAGCTGGGTC (N18-25) Gene Specific Seq 3’

Forward and reverse primer pairs (Illumina/Invitrogen, Carlsbad, CA) at a concen-


tration of 25 µM are combined into a master mix containing 0.15 µM of each dNTP,
reaction buffer and 40 ng of genomic DNA with total reaction volume of 35 µl.
Typical cycling conditions after a 1 minute initial denaturation at 98°C are as follows:
98°C for 10 seconds, 55°C for 30 seconds and 72°C for 1 minute per kb intended
product size. Reactions are cycled through these temperatures 25 times followed by
a 72°C final extension of 10 minutes. After cycling, the PCR products are transferred
to 384-well Millipore filter plates (Millipore, Billerica, MA) using a Beckman
Coulter (Beckman Coulter, Fullerton, CA) Biomek FX 96 probe liquid handling
robot. Filter plates are then subjected to a vacuum of 10 inches of Hg for approxi-
mately 10 minutes. Then, 50 µl of Milli-Q water is added to each filter plate, PCR
products are eluted by aspiration and then transferred to a clean, 384-well MJ
Research hardshell plate (Bio-Rad Waltham, MA) using a Beckman Coulter Biomek
FX 96 probe liquid handling robot.

PCR PRODUCT VERIFICATION AND QUANTITATION


Each PCR product is analyzed by capillary electrophoresis on the Caliper Life-
Sciences (Hopkinton, MA) AMS 90 SE Instrument using LabChip HT 2.4.1 software.
A 384-well or 96-well plate containing PCR sample volumes no less than 25 µl is
placed in the instrument. Two trays containing ladder and buffer respectively are
equipped alongside a “Caliper Chip” which must be cleaned, primed, and prepared
with gel dye and marker. Before beginning a run, an input file, containing only ORF
IDs and ORF lengths are loaded into the computer along with user input of the plate
type and allowed percent deviation from actual ORF length. This information will
be used to later calculate the concentration of the fragments found. A 96-well plate
will take about an hour to resolve; consequently a 384-well plate will resolve within
4 hours.
After finishing the run, two files are generated and saved. One output file contains
the actual electronic gel images and the second output file (of entirely text format)
contains the summary data obtained from every well including concentration, size,
and if the original fragment was found. This second output file is converted by virtue
9809_C001.fm Page 20 Friday, January 12, 2007 3:01 PM

20 Functional Protein Microarrays in Drug Discovery

of a simple script into a .csv file and is directly imported into the Biomek FX Span-8
liquid handling robot for automated set up of BP cloning reactions using equimolar
quantities (50 fmol) of target vector and PCR product insert.

BP CLONASE REACTIONS
BP cloning reactions are performed in 96-well, MJ Research (Bio-Rad Waltham,
MA) plates and are conducted in a 15 µl total volume containing: 50 fmol entry
clone vector, pDONR221, 50 fmol PCR product, 2 µl of proprietary BP clonase
enzyme, (Invitrogen, Carlsbad, CA) 3 µl of 5x BP clonase buffer (Invitrogen,
Carlsbad, CA) and brought to volume with 1 × TE. Reaction plates are then incubated
for 16 hours at 25°C in a thermal cycler, followed by 4°C hold until recovered. BP
clonase reactions are terminated through the addition of 2 µg of proteinase K
(Invitrogen, Carlsbad, CA) for 20 minutes at 37°C.

DH10B-T1 E. COLI TRANSFORMATION


Chemically competent, DH10B-T1 E. coli cells in 96-well format (Invitrogen, Carls-
bad, CA) are thawed on ice and 2 µl of the BP cloning reaction are added using the
Biomek FX 96 probe instrument. The plates are sealed with sterile covers and
incubated on ice for 30 min. The plates are transferred to a thermal cycler, prewarmed
to 45°C. The plates are held for 30 seconds and immediately transferred to ice for
2 min. Cells are allowed to recover by adding 40 µl of SOC media and incubating
at 37°C without shaking for 1 hour. Qtray bioassay trays (Genetix Limited, U.K.)
with 48 divided areas containing LB media supplemented with 50 µg/ml kanamycin
and 2% agar are warmed to room temperature. Several 3 mm glass beads are added
to each well and 30 µl of cells are then pipetted onto the agar surface. The Qtrays
are shaken gently until all visible liquid has been absorbed into the plates. The beads
are discarded and the plates are incubated at 37°C for 16 to 18 hours. Transformation
efficiencies are scored by colony count estimations (1–10, 10–50, >50) for each
transformation. The Qtrays are held at 4°C.

CLONE SEQUENCE VALIDATION


Colonies are picked with sterile toothpicks into 1250 µl of 2x YT media, supple-
mented with kanamycin 50 µg/ml, in 96 deep well blocks. The blocks are sealed
with an airpore tape pad strip and incubated at 37°C for 17 hours (11 hours static
and 6 hours shaking at 800 rpm). These inoculations are performed in duplicate,
one being specified for stock generation. Freezer copies of clone sets are generated
in 96-well Matrix Track Mate 2-D bar-coded vials (Matrix Technologies Hudson,
NH) by combining 50 µl of overnight culture to an equal volume of 75% glycerol
using the Biomek FX 96 probe liquid handling robot.

PLASMID EXTRACTION
Plasmid DNA is purified using what is essentially the Qiagen (Qiagen, Valencia,
CA) R.E.A.L preparation method using Qiagen’s QIAfilterTM and appropriate buffers.
9809_C001.fm Page 21 Friday, January 12, 2007 3:01 PM

High-Throughput Gene Cloning Using the Gateway® Technology 21

E. coli overnight cultures are collected by centrifugation of deep-well blocks at 3200


rpm for 15 minutes at 4°C. The growth medium is decanted and the pellets are
resuspended in 300 µl of R1 buffer (50 mM Tris pH 8) containing a final RNase
concentration of 100 ng/ml. The cells are lysed by the addition of 300 µl of R2
buffer (1% [w/v] SDS, 200 mM NaOH) with gentle mixing. The lysates are incubated
at room temperature for 5 min. The lysates are neutralized by the addition of 300
µl of R3 buffer (3 M KOAc, pH 5.5) followed by mixing and incubation on ice for
10 min. The Biomek FX 96 probe instrument is used to transfer lysates into QIAfilter
TM filter plates. The lysates are cleared by vacuum filtration. The plasmid DNA is
precipitated through the addition of 625 µl of isopropanol. After mixing, the plates
are spun at 3200 rpm for 30 min. at 4°C. The supernatants are decanted and the
pellets are washed with 300 µl 70% (v/v) ethanol (–20°C). The plates are spun at
3200 rpm for 15 minutes. The supernatants are decanted and allowed to air dry to
completion. The plasmids are resuspended in 50 µl of Blue Tris dye (1 mM Tris pH
8.0, bromophenol blue 1.25 mg/ml) by shaking for 30 minutes on a platform shaker.

SEQUENCING TEMPLATE PRODUCTION BY TEMPLIPHI


Sequencing templates generated by TempliPhi (Amersham Biosciences, U.K.) uses
the Phi29 DNA polymerase and rolling cycle amplification to generate linear
concatenated copies of plasmid templates. The Biomek 96 probe instrument is used
to transfer 10 µl of overnight culture into 384-well plates. The cells are collected
by centrifugation at 3200 rpm for 5 minutes. The media is decanted by inverting
the plates on absorbent material followed by low-speed centrifugation at 500 rpm
for 2 to 3 sec. The Biomek Span-8 is used to add 2 µl of lysis buffer to cell
pellets/well. The plates are then placed in a thermal cycler and incubated at 93°C
for 3 min. The plates are returned to the Biomek and 34 µl of Milli-Q H2O is added
to each well. The plates are then sealed and spun at 3200 rpm for 5 min. Two
microliters of the supernatant are transferred to a clean 384-well MJ Research plate
(Bio-Rad Waltham, MA). The enzyme (4 µl) is added to the supernatants and the
plates are sealed and incubated at 30°C for 16 hours. The reactions are stopped by
heat treatment at 96°C for 5 min. The Biomek FX 96 probe liquid handling robot
then adds 34 µl of Milli-Q H2O Blue Tris dye to each well.

REFERENCES
1. Marsischky, G. and LaBaer, J., DNA many paths to many clones: A comparative look
at high-throughput cloning methods, Genome Res., 14, 2020, 2004.
2. Hartley, J.L., Temple, G.F., and Brasch, M.A., DNA cloning using in vitro site-specific
recombination, Genome Res., 10, 1788, 2000.
3. Esposito, D., Gillette, W.K., and Hartley, J.L., Blocking oligonucleotides improve
sequencing through inverted repeats, Biotechniques, 35, 914, 2003.
4. Churchill, G.A. and Waterman, M.S., The accuracy of DNA sequences: Estimating
sequence quality, Genomics, 14, 89, 1992.
9809_C001.fm Page 22 Friday, January 12, 2007 3:01 PM
9809_C002.fm Page 23 Friday, January 12, 2007 3:04 PM

2 Protein Expression
for MicroArrays
Harry H. Yim, Thomas G. Chappell,
and Steven H. Harwood

CONTENTS

Introduction..............................................................................................................23
E. coli Expression....................................................................................................24
Introduction..................................................................................................24
Optimizing Soluble Protein Expression ......................................................24
HTP Methods for Detecting Protein Solubility ..........................................26
Protein Purification ......................................................................................27
Yeast Expression......................................................................................................27
Introduction..................................................................................................27
Early Heterologous Protein Expression in Yeast ........................................28
Pichia pastoris Expression Systems ............................................................29
Posttranslational Modifications ...................................................................29
HTP Yeast Expression .................................................................................30
Insect Cell Expression .............................................................................................31
Introduction..................................................................................................31
Cloning for Insect Cell Expression .............................................................31
Posttranslational Modifications ...................................................................32
HTP Baculovirus Expression ......................................................................33
Choosing Expression Technologies for a HTP Pipeline.........................................34
References................................................................................................................35

INTRODUCTION
With the completed sequence of the human genome, as well as the sequencing of
the genomes of hundreds of other species, the structure and function of literally
hundreds of thousands of proteins are of potential interest in diverse fields of
biology. Proteomic studies increasingly require the expression of large numbers
of proteins in parallel. No one expression system has proven be ideal for all types
of downstream applications; each host having advantages and disadvantages when
evaluated for protein yield, functionality, posttranslational modifications, high-
throughput (HTP) capacity and cost. Trade-offs are required to optimize high-throughput

23
9809_C002.fm Page 24 Friday, January 12, 2007 3:04 PM

24 Functional Protein Microarrays in Drug Discovery

output based on downstream requirements, which are frequently mutually exclusive.


Posttranslational modifications, for instance, might be critically important in drug
screening applications but incompatible with crystallization studies. As a result,
open reading frames are often expressed in a variety of heterologous host systems;
an approach that has been facilitated by the development of improved cloning
methods such as the Gateway® system, which utilizes in vitro recombination,1
allowing the rapid and flexible cloning of recombinant DNA into a variety of
expression vectors.

E. COLI EXPRESSION
INTRODUCTION
E. coli remains the most widely used system for rapidly expressing large numbers
of proteins and is in many respects a model system for high-throughput protein
production. Protein expression in E. coli is relatively reliable, robust, simple, amenable
to HTP expression, and cost-effective. Continual improvements have resulted in
increased throughput and decreased growth volumes. There are also well developed
protocols for cloning, expression and purification, many which have been highly
optimized and automated for small scale.2,3 Despite the advantages that the E. coli
protein expression system provides, there are some drawbacks to using E. coli
as an expression host. These include lack of posttranslational modifications and
contamination of protein product with endotoxin. However, the most significant
problem for proteomics applications is that proteins expressed in E. coli often
accumulate as insoluble and inactive aggregates. One possibility for the high fraction
of insoluble proteins may be related to the use of the popular and well established
T7 expression systems. In this approach, one employs the bacteriophage T7 late
promoter on medium copy number plasmids. The highly active T7 RNA polymerase
is provided by the host cell and regulated by the IPTG-inducible lacUV5 promoter.
While this system provides very high concentrations of recombinant protein, it may
be a victim of its own success in that it may produce more protein than the cell is
capable of properly folding.

OPTIMIZING SOLUBLE PROTEIN EXPRESSION


One common approach to improving solubility for T7 and other systems is to alter
the expression conditions to promote folding. These are generally applicable for
HTP format and are geared towards reducing the rate of protein synthesis to provide
more time for folding. Methods to decrease the expression levels include using lower
concentrations of IPTG or coexpression of phage T7 lysozyme (which degrades T7
RNA polymerase) from compatible pLysS and pLysE plasmids. Another approach
is to use a promoter with the native E. coli RNA polymerase, rather than T7 RNA
polymerase, to transcribe the mRNA. This allows more efficient coupling of tran-
scription and translation, potentially leading to more soluble product. Another easy
and effective method is to reduce the growth temperature and allow a longer period
for protein synthesis. For example, rather than performing the expressions at 37°C
for 3 hours, temperatures are reduced to between 18°C and 30°C and proteins
Exploring the Variety of Random
Documents with Different Content
BRAND!

Den negenden dag, dat de N i e u w - H o o r n voor Sante-Marie lag, waren allen


genezen. En men ging welgemoed onder zeil, vertrouwende, dat de voorraad
versch voedsel toereikend zou zijn. Men koerste eerst Zuid-Oostwaarts tot op 33°
en wendde den steven daarna Noord-Oost naar Straat Soenda.

Het waren mooie, stille dagen. De maats hadden de handen vol met het
onderhoud van het pluimvee. Maar heel hun leven hadden de oomes niet zooveel
eitjes gepeuzeld.

Padde werd in den loop der weken zelf zoo rond als een ei. Men ried hem wat
beweging aan. Zuchtend besloot Padde het werkje over te nemen, dat tot nu toe
de Schele altijd had verricht, namelijk ’s middags met een vaatje in den kelder te
gaan en dat vol te pompen, ten einde den volgenden morgen allen oomes een half
„mutseke” te kunnen verstrekken.

Joppie kreeg spreeklessen. Een kreet van blijde verrassing ging onder de oomes
op, toen hij duidelijk verstaanbaar: Hajo! krijschte. Maar nu bleek, dat Joppie al
niet veel beter was dan de menschen: ook hij stelde zijn kennis in dienst van het
booze. Hij riep zijn meester den ganschen dag, liefst barsch, bevelend, zooals
Berentsz. het deed, dan weer lokkend, vleiend, of angstig, opgewonden, als wilde
hij zeggen: Kerel, je bent me toch niet overboord gevallen?—Zoo kwam Hajo soms buiten adem
aanhollen om te vragen wat de bootsman van hem wou, en vond dan, in plaats van een grimmigen
Folkert Berentsz., een allervriendelijksten Joppie, die hem den kop toestak om gekrauwd te worden.
Alle maats, die den leergierigen vogel hun naam wisten in te pompen, kregen er spijt van als haren
op het hoofd. [228]

Joppie leerde ook zijn eigen naam en toonde in het uitspreken ervan een groote mate van
zelfingenomenheid. De maats wisten niet waar ze bleven van het lachen, wanneer Joppie zijn naam
in alle toonaarden, den een nog vleiender en liefelijker dan de andere, uitgalmde.

Gerrit was tevreden over zijn pleegkind. Hij had aanvankelijk wel wat raar tegen den krommen snavel
en de bonte pluimage van den jongen „torenkraai” aangekeken, maar nu herkende hij in de wijze,
waarop Joppie: Ka! kon zeggen, toch duidelijk een rasgenoot.

Toen de oomes vonden, dat Joppie meer dan wijs genoeg was, nam Padde de taak over om Joppie’s
leergierigheid te bevredigen.

„Vooruit!” zei Padde, „zeg nou eens: Padde Kelemeijn!”

Joppie keek hem pienter aan. „’t Is een merakel!” meende het dier toen.

„Vooruit!” mopperde zijn leermeester. „Padde Kelemeijn! Zeg het dan, stommeling!”

Joppie hield z’n kop schuin, luisterde vol aandacht. Toen keek hij Padde trouwhartig aan en
schetterde: „Stommeling!”
„Je bent zelf een stommeling!” gromde Padde, die rood werd van drift.

„Kletskoek”, verklaarde de vogel en keek luchthartig naar boven.

Padde staarde het dier met opengespalkte oogen sprakeloos aan, keerde zich toen om en nam zich
voor, geen woord meer aan het mormel te verspillen.

Zoo kwam een dag, die den mannen van de N i e u w - H o o r n lang heugen zou: de negentiende
November van het jaar 1619.

Naar gewoonte begaf Padde zich in den namiddag naar den kelder, daalde met zijn kaars het korte
trapje af en plaatste het licht op een volle ton, om de handen vrij te hebben voor het pompen.

„Mallemallemootje,
Zeven in een bootje;
Mallemallemootje, mallemallemoer,
Zes aan de riemen en een aan het roer!”

[229]

zong Padde, blijgemoed pompende. Het vaatje was vol; Padde bevrijdde met zwierigen greep de
kaars, die hij op de ton had vastgesmolten. Toen viel—kon het ongelukkiger?—het gloeiend eindje
der kaarsepit in het spongat; de brandewijn daarbinnen vatte vuur; de duigen scheurden met een
doffen knal uiteen, en de brandende vloeistof dekte ineens den geheelen kelderbodem. Met een
schreeuw vloog onze botteliersmaat het laddertje op, zag twee putsen water staan, waarmee Hajo en
Rolf aan het dekschrobben waren, greep de putsen en keerde ze boven het luik uit.

„Padde! Wat is er?!”

De arme dikzak wilde wat stamelen, maar het was niet meer noodig: het sissen, knetteren en de
wolk verdampt water, die uit het luik opsloeg, zeiden genoeg.

„Brand! Brand!!!”

Dat werkte. Van alle kanten kwamen de maats met verschrikte gezichten aanhollen, sommigen al
met volle putsen. „Wáár is de brand?!”

„In de kelder!!” Angst en opwinding trilden door aller stem. In razende haast zocht men naar putsen.
Heele plassen water werden door het luik geworpen.

Folkert Berentsz. klauterde, nadat men wel een honderd emmers water had leeggeworpen, den
kelder in en kon daar gelukkig geen brand meer ontdekken. Intusschen had Bontekoe zich naar het
ruim gehaast, waar, zooals hij terecht vermoedde, plassen brandend vocht den bodem dekten. Hij
riep om putsen. Na een tiental te hebben leeggegoten, scheen ook daar de brand gebluscht. [230]

Een zucht van verlichting ging op, toen men hoorde, dat de smidskolen niet door het vuur waren
aangetast. Nog hijgend van opwinding, besprak men het gevaar, dat gedreigd had. Als de brand tot
in den kruitkelder was doorgedrongen.….

„Padde!” werd er geroepen. „In de kajuit komen!”


De arme jongen trilde over al zijn leden. Ongemerkt wist hij
door een luik in het ruim te glippen. Daar was het stikdonker;
tastend daalde hij het steile trapje af. Beneden gekomen,
wankelde hij naar een hoop touw, ging zitten, borg het hoofd
in de handen en schreide, schreide.….

Stil! Wat hoorde hij daar?! Met bonzend hart luisterde Padde.
Hij durfde de oogen haast niet te openen in dit griezelig
donker. Krak! Knaps! Kritsch! Klappertandend richtte Padde het
hoofd op. Was het een koortsschim? Daar, aan den anderen
kant van het ruim, laaiden groote vlammen op! Een zwavelige
lucht drong Padde in den neus. Groote God.….! De kolen
brandden.….!!

Met een schreeuw vloog Padde naar de ladder. „De kolen …! De


k-kolen!!!

Een nieuwe paniek. „De kolen?! Branden de kolen?!!”

Een paar kloeke maats zijn het eerst weer het ruim in,
gewapend met putsen water, die ze hijgend leegkletsen over de
brandende kolen. Gesis van heb-ik-jou-daar; gele
zwaveldampen barsten uit den gloeienden berg te voorschijn,
en in een oogwenk is de lucht in het ruim zoo benauwend, dat
men het er geen twee minuten kan uithouden.

Maar de kerels zijn taai. Altijd opnieuw dalen ze met volle


putsen langs het smalle laddertje in het ruim af, banen zich een
weg in het verpeste zwavelhok, werpen het water over de
kolen en zoeken tuimelend, vloekend, met betraande oogen
den weg terug naar het laddertje. Hoevelen vinden den weg?
Hoevelen zakken bedwelmd, reeds half verstikt ineen, na in
radeloozen angst heen en weer gerend te zijn?

Bontekoe leidt zelf het werk,—tot zijn stem versmoort, en hij


wankelend de ladder op vlucht. Maar een minuut later is hij
weer beneden. „Moed, jongens!”
Naar gewoonte begaf Padde zich.…
Men kapt gaten in het tusschendek, werpt ontzaggelijke
hoeveelheden water in het ruim. Zou het helpen? De planken [231]bodem onder de voeten wordt
steeds heeter; de maats springen als zandvlooien rond, moeten om den haverklap hun half
verschroeide voetzolen koelen in de putsen water, die ze aandragen.

Zou men het kruit overboord werpen? De kans om in deze streken een Spaansch schip te ontmoeten
is bedenkelijk groot. Zonder kruit aan boord zou men verloren zijn. Nu, met het wegwerpen van het
kruit dan maar tot het uiterste gewacht.

Hou vast, mannen!


Maar er waren verraders. Wetende, dat de jol en de sloep achter het schip aansleepten, (de jol was
sinds het vertrek van Sante-Marie nog niet binnengehaald, en de sloep was daareven uitgezet, omdat
ze bij het bluschwerk in den weg stond) hadden enkelen het voor raadzaam gehouden zich overboord
te laten glijden, naar de jol of de sloep te zwemmen en zich onder de banken te verbergen. De
koopman, die juist naar de kajuit ging om in elk geval reeds zijn voornaamste papieren bijeen te
binden, zag juist een kerel de jol inkruipen. „Wat heeft dat te beteekenen!” schreeuwde hij.

„Kom ook in de booten, Koopman!” riepen de maats. „Temet vliegt de heele kast aan gruzelementen!”

„Als jullie niet terugkomt, waarschuw ik de schipper!” riep Hein Rol driftig.

„Dan kappen wij de touwen door!”

„Schurken!” was Rol’s verontwaardigd antwoord.

„Ga je mee of niet?” werd er uit de jol geroepen.

De koopman weifelde even, maakte een onwillig gebaar. „Ik kom!” riep hij toen norsch. En hij
spoedde zich naar de kajuit om zijn paperassen.

Want kostbaarder dan zijn eer, waren hem zijn papieren.

Toen hij zich langs een touw in de jol had laten glijden, kapten de maats de booten vrij. „Roeien jullie
weg?!” vroeg Rol verschrikt.

„Neen, waarachtig niet, we zullen in de buurt blijven om te helpen. Maar zoo meteen vliegt de boel
aan stukken, dan moeten wij buiten schot zijn.”

De koopman zweeg, keek met bezorgd gelaat naar het schip, waaruit een vuil-gele rook opwervelde,
die masten en touwen [232]aan het oog onttrok. „’t Is gekkenwerk om nog aan blusschen te denken!”
zei hij, om zijn geweten gerust te stellen. En hij blies zich in de bleeke handen, die geschaafd waren
door het omlaagglijden langs het touw.

Op het schip vochten de anderen. Met toegeknepen oogen gaven de mannen elkaar de volle putsen
over. Hou vast, jongens!!

Daar kwam de barbier aanhollen. „Schipper! De booten zijn weg!!”

De mannen zijn verlamd van schrik. „De booten weg?!!” Alles vliegt naar de verschansing. Daar
drijven de booten!! Een stomme woede maakt zich van de verlatenen meester.

„Schipper! Wat nou??!”

Zóó hebben de mannen hun schipper nog nooit gezien. Het open zeemansgelaat is plots vertrokken
van toorn en smart. „Haal de zeilen om, mannen! We zullen ze onder de kiel stroopen!”

Smart over de wonde, die hun makkers hun geslagen hebben, doet de maats het want invliegen en,
tastend in den groezeligen rook, met opeengeklemde tanden de zeilen krap zetten. Zoo koerst men
recht op de booten af.

Daar schijnt het dreigend gevaar vermoed te worden. De kerels roeien als dollen, steken drie
scheepslengten voor de N i e u w - H o o r n over, den boeg in den wind, zoodat ze niet achtervolgd
kunnen worden.

„Wel, laat hun geweten hen dan straffen!” roept de schipper. „Nu het kruit maar overboord, mannen!
De schuit drijft nog! Als we er aangaan,—dan met z’n allen!”

„Leve Bontekoe!!!” brullen de oomes, al is het alleen maar om den laffen kameraden daarginds te
laten hooren, dat er nog kerels zijn, die niet in de booten kruipen, als de schuit in gevaar is. Grimmig
pakken ze met hun verweerde knuisten de tonnetjes kruit, werpen ze van man tot man en zoo
overboord. Op Bontekoe’s bevel laten de maats, die met timmergerei weten om te gaan, zich over de
verschansing zakken, om onder het zee-oppervlak gaten in den scheepswand te boren. De schipper
wil het ruim een paar vadem vol laten loopen en zoo den brand [233]van onderen blusschen. Maar
vergeefs zet men de boren in het hout: de wand zit vol ijzerwerk. Dan maar weer met putsen aan het
werk!

Hoei!! De vlammentongen lekken al uit een der luiken. Dat is de duivel, die er uit loert! Smijt hem
den kop in!

In koortsige opwinding komen de mannen met water aandragen.

Maar hoogerop laait de vlam, en vreemd.…. het kraken, knetteren, barsten, piepen houdt op.
Zachtjes loeiend steekt het vuur een lange, roode tong uit de luiken.

„De olie brandt!!!”

Met verlamming geslagen laten de oomes de armen zinken; dikke tranen rollen den gebruinden
kerels over de wangen.

„Water!” roept een stem.

Flang! daar gaan de putsen weer rond. Al gietende en de voeten koelende, grienen de oomes als
kinderen. Maar opgeven? Ho maar! Hollandsche jongens, wâ-blief?! Grienen kan geen kwaad: tranen
zijn ook water en helpen blusschen. Met hun bloote pooten staan ze schrap op het gloeiend heete
dek.

Een der dapperste voorwerkers, die koppig, oogen en lippen saamgeperst, in rook en damp te
ploeteren stond.…. was Padde.

Hij boette zijn schuld.

En men werkte. De wanhoop in het hart,—maar men werkte. Tot.….! Met verdoovend gekraak begaf
zich het achterdek, en gillend stortte een handvol wakkere vrienden in de vuurzee. Hei!!! Hoe stoven
de vonken tot hoog boven de masten uit! En de vlammen sloegen tegen de ra’s op, grepen de zeilen
aan en zonden de blanke vleugels van de N i e u w - H o o r n verzengd, wijd uitlaaiend omhoog. De
gescheurde kabels vielen slap neer in den vuurgloed en dienden duizend kleine vlammetjes tot
ladder.

De vlag moest veroverd worden! Op, vlammen! Haalt het neer, dat dartele, bonte doekske!

Daar reet een vlam het omlaag en roofde het zijn kleuren. Danst nu, vlammen! Danst; de vlag is ons,
danst, danst! [234]
Hajo, Rolf en Padde stonden bij den grooten mast. Verderop begon het dek door te buigen; er
schoten bruine schroeivlekken in.…. Toen sleurde Rolf zijn beide makkers naar de verschansing. „Het
water in!” siste hij tusschen de tanden. De verbouwereerde knapen volgden klakkeloos het bevel op.

Velen waren Rolf nagesneld en ook pardoes in zee gesprongen. Anderen hadden, verlamd van schrik,
niet zoo snel een besluit kunnen vatten en zonken nu weg in de vlammen.

Toen gebeurde het. Een ontzettende slag, een helsch gekraak, een prikkelend-scherpe lucht,
versmoorde kreten, angstig geloei van het vee.…. Het kruit had vlam gevat.

Een nieuwe ontploffing! De N i e u w - H o o r n scheurde uiteen; masten, planken, menschen, dieren


en brokstukken ervan vlogen de lucht in. Sissend, krakend, hoog opstuwend de kolommen zwarte
rook en gouden vonken, kantelden de brokstukken van het gescheurde schip weer tegen elkaar
en.…. zonken in de golven weg.

De mannen in de boot keken rillend toe.

Was het mogelijk?! Was de N i e u w - H o o r n , hun prachtig schip, vergaan?! Die zwarte wolk tegen
den bloedrooden avondhemel.…. was dat alles, wat er van restte?!—Neen! Daar dreven op het water
stukken mast, kisten, balken, en op die stukken mast, die kisten en balken klemden zich levende
wezens. Te hulp! De handen aan de riemen!!

Van onze drie Hoornsche vrienden was het Hajo, die het eerst zijn bezinning terugvond. Hij zag den
bezaansmast drijven en werkte zich er op. Snel denkend en handelend, wierp hij Padde, die zich aan
een langzaam volloopende houten bak had vastgeklemd, een masttouw toe. Padde zag het, tastte er
naar, maar vond het niet. Hajo trok het touw in en wierp het opnieuw. Ditmaal wist Padde het te
grijpen, liet zich naar zijn makker trekken en pakte hem kreunend bij de knieën. „Hajo.…. o, God,
Hajo.….!”

„Klim op de mast!”

„Ik kan niet meer.….!”

Met inspanning van alle krachten wist Hajo zijn vriend schrijlings op den mast te krijgen. Snikkend
leunde Padde het [235]hoofd tegen Hajo’s schouder.—Nu Rolf! Groote God, waar zou Rolf zijn! In
radeloozen angst keek Hajo rond. Verderop, buiten zijn bereik, worstelden een paar maats met de
golven. Hier en daar werkten zich in de schemering gestalten op stengen, planken, tonnen of
brokken mast. „Rolf! Rolf!! Rolf!!!”

„Hajo!”

Goddank! Rolf had zich op een mastkorf gered.—Waar waren de booten? Te ver om ze te beroepen.
In de halve duisternis was het niet mogelijk te zien, of ze hier vandaan, dan wel hier naar toe
roeiden. „Boot ahoy! Ahoy!”

„Hajo.….!” kermde Padde.

„Moed, Padde!” Hajo sloot zelf even de oogen om wat rust te vinden. Langzaam vloeiden de tranen
over zijn wangen. Toen hij de oogen opsloeg, was de jol tot op een vijftig voet genaderd. Dichterbij
kon ze niet komen door de vele, zware brokstukken, die overal ronddreven. Hajo mat den afstand tot
de jol. Zou hij het halen? „Blijf hier zitten, Padde! Hou je goed vast!”
„O, God, Hajo.…. je gaat toch niet weg? Hajo.….?!”

Hajo beet de lippen opeen, liet zich van den mast glijden en zwom in de richting van de jol. Maar
onderweg werden zijn armen zwaar als lood.….

Daar werd hem een touw toegeworpen. Hij greep het en liet zich naar de jol trekken. „Neen!” hijgde
hij, toen de oomes hem binnen boord wilden halen, „ik rust maar even. Geef me het.…. het touw
mee.…. ik wil.…. ik.….” Zijn oogen sloten zich; zijn handen lieten den jolboord los; men kon hem nog
net bijtijds grijpen en binnen boord halen.

Toen sprong Bokje, de trompetter, met een loodlijn overboord, zwom naar Padde en liet zich samen
met hem terugtrekken. Een tweede maat ging het water in en redde den Neus. Floorke kwam
proestend, op eigen kracht aanplassen en begon den kerels in de jol de huid vol te schelden. Maar
toen hij ze een voor een hun huid zag wagen om den haaien een drenkeling te ontrooven, zakte zijn
gramschap weer een weinig. Men roeide de plaats rond, waar de N i e u w - H o o r n was
ondergegaan. Ten slotte geloofde men alle drenkelingen te hebben opgepikt. Velen verloren het
bewustzijn terstond, wanneer ze [236]in de boot waren getrokken; anderen lagen nog van opwinding
te schreien. Maar de schipper?! Waar was de schipper!!

Daar verscheen Harmen’s kop aan bakboord. Hij liet zich in de jol trekken, spuwde een golf zeewater
uit, wees verderop. „De sch.…. schipper!” Toen begaven zich zijn krachten.

Men zag in de door Harmen aangeduide richting in het duister een wrakstuk drijven en daarop een
gedaante. De jol werd zoo dicht mogelijk naar den drenkeling toegeroeid; Bokje, de beste zwemmer
van allen, sprong weer met een lijn overboord, en, ja, even later kwam hij met den schipper terug,
die in de jol getild en achter in de roef werd neergelegd. Goddank! „Geef ons raad, schipper! Wat
moeten we doen?!”

Met matte stem ried Bontekoe aan, dezen nacht nog bij het wrak te blijven en den volgenden morgen
wat levensmiddelen op te hengelen, die overal ronddreven. Hevige pijnen deden hem spoedig het
bewustzijn verliezen. Men roeide nog eens om de ongeluksplek heen, maar vond in het duister geen
menschelijk wezen, dood of levend, meer. Toen haalde men de riemen in, om den morgen af te
wachten.

Maar met de verschrikking voor oogen valt wachten moeilijk. Zij, die van vermoeidheid waren
ingeslapen, werden met een gevoel van onrust weer wakker.

„Laten we wegroeien!” zeiden ze. „Waarom roeien we niet weg?”

De anderen schudden het hoofd. „We moeten morgen wat eten opvisschen. Met het beetje brood,
dat we hebben, houden we het geen dag uit.”

Maar de vragers waren niet tevredengesteld. „Wat hebben we aan eten, als de zee gaat aanloopen
en de jol aan stukken slaat? Nu is het goed weer; laten we naar land zoeken!”

„We moeten wachten! Bevel van de schipper!”

Dan werd er een tijdje gezwegen.—Maar een nacht is lang. Een kwartier later begon de onrust weer.
„Laten we toch wegroeien! We zullen wel een paar dagen vasten! Misschien hebben we morgen al
land. We zijn immers niet ver meer van Sumatra!”
„We moeten wachten. De schipper heeft het gezeid.”

„Nou ja.….” [237]

„Wat: nou ja?! Ben jij soms een van die gluiperds, die er tusschen uit zijn gegaan?”

„Als wij dat niet gedaan hadden, waren jullie met z’n allen voor de haaien geweest!”

„Toch was het smerig!”

„Weet je, wat i k smerig vind? Dat jullie er ons onderdoor wouden halen!”

„Dat was je verdiende loon geweest!”

„Maar het zat jullie toch niet glad, hè?”

„Kom”, zeurt een ander, „maak geen herrie! Laten we nou wegroeien!”

„Neen”, koppen een paar oomes, „de schipper heeft gezegd: neen.”

Nieuw zwijgen. De minuten kruipen.

Ineens vloekt er een oome, steekt de riemen uit en begint te roeien. En daarmee is de ban
verbroken, die van ’s schippers woord uitging. Allen, die nog macht over hun lichaam hebben, grijpen
een riem.

Waarheen? Naar land! Waar ligt dat? Niemand, die het weet. Maar het roeien drukt de onrust den
kop in. Roeien, jongens! Roeien! Niemand houdt het stuur.

Enkelen worden wakker, ontwakend uit een nachtmerrie. „Waar zijn we?!”

„In de jol.”

„In de jol?” Even zwijgen. De vrager schijnt zijn herinnering wakker te roepen. „En waar gaan we nou
heen?”

„Naar Sumatra.”

„Waar legt dat?”

„Vlak bij. Pas maar op, straks val je er nog over!”

Zwijgen. Stug roeien de kerels voort.

„Hein”, vraagt er een met zwakke stem, „was jij dat, Hein?”

„Ja, Kalle. Waar lig je?”

„Voorin.—’k Heb zoo’n pijn.….”

„’t Zal wel klaren, Kalle. Ik heb een verbrande poot.”

„Hou hem in het water.—Zouwen we ver van land zijn, Hein?” [238]
„Ben je gek? Morgen, als het licht is, zien we ’t misschien wel. Wacht maar, Kalle, als ’t licht is,
morgen.….!”

De roeiers zuchten, halen de riemen aan.

Platsch!—Platsch!—Platsch!

[239]
[Inhoud]
IN DE BOOTEN
Eindelijk klaart de morgen. Men tracht met de moede oogen den ochtendnevel te
doorboren, die over het water hangt. Nergens land te zien.….—Ook de sloep is uit het
gezicht verdwenen.

Grienend laten de oomes de riemen zinken. Nu eerst voelen allen hoe afgemat ze zijn. Als
de zon opgaat, is er geen een meer wakker.

Stuurloos dobbert de jol op den stillen golfslag. Een heerlijk blauw uitspansel welft zich
boven de onafzienbare watermassa.

In den middag ontwaken enkelen. De slaap heeft verkwikking geschonken; de kerels


voelen hun hoop weer opleven: de schipper is aan boord en zal wel goeden raad geven.
Fluisterend, als waren ze in de stilte rondom voor hun eigen stemmen bang, bespreken ze
den ondergang van hun prachtige schuit en het verlies van al hun schatten. De een had
nog vijf vette ganzen gehad; die dobberden nou stellig ook ergens rond! Een ander had
zijn mes verloren, z’n puike, fijne messie, dat ie verleden Zaterdag nog zoo lekker had
aangezet. Zou je niet spinnijdig worden op dien beroerden botteliersmaat met z’n kaarsje?
En z’n vrind Nelis was ook naar de weerlicht. Z’n vrind Nelis, waarmee-d-ie al wel door
een dozijn schipbreuken goed was heengerold. Z’n mes en z’n vrind weg!

Allengs werden allen wakker. Men wekte den schipper. „Wat [240]moeten we doen,
schipper? We zien het wrak niet meer en ook geen land.”

„Zijn jullie dan toch van het wrak weggeroeid?”

„Ja, schipper, we dachten.….”

„Dat was verkeerd. Is er een zeil in de jol?”

Men zocht onder de plecht en de banken. „Neen, schipper, geen stukkie zeil.”

„Trek dan de hemden uit en maak er een zeil van.”

Men sloeg vol vertrouwen aan het werk. De schipper zou hen wel naar Sumatra brengen!
De stootballen werden binnen boord gehaald en tot garen uitgeplozen. Toen men
voldoende meende te hebben, trokken de maats de hemden uit en begonnen ze aaneen
te naaien tot twee zeilen.

De barbier ging den toestand der gewonden na. Haast allen hadden zich min of meer de
voetzolen verbrand. Eerst werd de schipper behandeld, die twee hoofdwonden had. Vader
Langjas kauwde iets van het weinige brood, dat de eerste vluchtelingen inderhaast
hadden meegesleept, tot een papje en legde dit op de kwetsuren. Ook de andere
gewonden werden aldus behandeld. Rolf had een brandwond aan het been, die
gedurende den nacht leelijk was opgeloopen en den braven barbier, die vaderlijke
gevoelens voor Rolf koesterde, bezorgd zijn grijzen bol deed schudden.

Men telde met z’n hoevelen men in de boot zat en kwam tot het getal zes-en-veertig. In
de sloep konden hoogstens tachtig man zijn geborgen. En dus de anderen.….!

Tegen de schemering waren de zeilen klaar. Men richtte den mast op, die in de jol lag, en
haakte er den boom en de gaffel in. Een opgestoken roeispaan diende voor fok. Toen de
beide masten stonden en de „zeilen” waren bevestigd, wendde men den steven Noord-
Oostelijk. De bedroevend kleine voorraad brood werd bijeengelegd, en allen kregen er
een vingerdikke snede van. Het was onrustbarend te zien, hoe zelfs het uitdeelen van een
zoo geringe hoeveelheid den voorraad slinken deed.

Padde had den heelen dag door geslapen. Toen hij ’s avonds ontwaakte van het lawaai,
dat met het oprichten van den mast gepaard ging, borg hij terstond het hoofd weer in de
armen weg en hield zich, angstig, slapende. [241]

Hajo was, als alle anderen, weer vol goeden moed, rekende er stellig op, dat, nu ze in de
goede richting zeilden, morgen wel land in ’t zicht zou komen.

Rolf, die de laatste paar maanden dagelijks de reis op de kaart gevolgd had, zag den
toestand minder rooskleurig in. De pijn aan zijn been stemde hem ook niet vroolijker en
maakte hem koortsig.

Zoo viel de duisternis in, ving alles in haar wijde armen.

Te middernacht maakt Gerretje een heidensch spektakel. „Land! Land!!” Alles vliegt
overeind. „Waar is land?!”

Aan bakboord, ver weg, pinkt een lichtje. Een dolle vreugde maakt zich van de
schipbreukelingen meester. Men gooit de zeilen om, grijpt naar de riemen. Dat licht kan
niet anders dan land zijn: midden op zee groeien geen lampjes als paddestoelen in een
wei, wâblief? ’t Zou een wallevisch kunnen wezen met een lichie op z’n knikker! Neen,
jongens, land is het! Floorke ziet al bergen. Morgen zullen ze onder de kokosboomen
wandelen. Roeien, jongens!!

Maar de bergen vervagen en stijgen als wolken omhoog. En in het lichtje komt beweging;
het schijnt op en neer te gaan.….! Enkelen weifelen in het roeien, als vreezen zij hun
angstig voorgevoel bewaarheid te zien. Dat lichtje daarginds is geen land! maar een
hulkje met schipbreukelingen.….

„De sloep.….!”
De schrijning der teleurstelling wordt verzacht door vreugde over
het weerzien. De sloep voert eveneens twee zeilen: lichtgrijze
vlekjes. Men gooit de riemen weer neer en wacht de sloep af;—
waarom verder van den juisten koers af te wijken?

„Sloep ahoy!”

„Ahoy!”

Namen van vrienden worden heen en weer geroepen. Kreten van


blijdschap, wanneer twee makkers elkaars stemmen herkennen.
[242]

„Hebben jullie eten?”

„Drie twee-ponds brooden. En jullie?”

„Niets.”

„Groote griebus! Welke koers varen jullie?”

„Heelemaal geen koers. En jullie?”

„Wij varen op de sterren. We hebben de schipper bij ons.”

„De schipper?! Hóór je, mannen, de schipper is in de jol!! Schipper,


ben jij daar? Leve de schipper, mannen!” Een schor, instemmend
gebrul. „Wanneer zullen we aan land zijn, schippertje? Morgen al?”

„Moed, mannen! Vertrouw op Gods Goedheid!”

„Amen”, zeggen een paar vrome oomes.

Gezamelijk werd de tocht voortgezet; de jol gaf de richting aan. Maar al spoedig bleek,
dat de sloep achterbleef. Men greep naar de riemen en haalde de jol weer in.
„Schippertje, neem ons over! We zullen mekaar verliezen! Neem ons over, dan zetten we
alle zeilen op de jol en varen nog ééns zoo vlug. Toe, schippertje.….”

Maar de oomes in de jol verzetten zich. „De jol is voor zooveel man te klein, schipper!” En
toen de maats uit de sloep zich aan den jolboord vastklemden, stieten de anderen de
sloep met ruw geweld terug.

Een jammerklacht steeg op onder de verschoppelingen. „Schipper! Niemand van ons kan
op de sterren varen! Moeten we dan om zeep gaan?”
Maar de oomes in de jol kenden geen erbarmen. „Als we jullie met z’n zes-en-twintigen
overnemen, zijn we allemaal voor de haaien!”

Zuchtend grepen de arme kerels naar de riemen. De olielantaren werd op de jol


overgebracht, zoodat men er zich in de sloep naar richten kon.

Langzaam werd het licht. De mannen tuurden naar alle zijden over het watervlak.

„Zie jij wat, Doedesz.?”

„Net zooveel als jullie.”

Allen zuchtten. Enkele oomes luchtten hun smart door Padde met verwijten te
overstelpen. De arme jongen begon te snikken, [243]en een paar anderen, in de eerste
plaats zijn beide vrienden, namen hem in bescherming.

Rolf voelde zich wat verlicht; zijn been stak hem veel minder, en met een gelukkig gezicht
legde Vader Langjas een nieuw papje op de wond. Ook de kwetsuren der anderen lieten
zich gunstig aanzien.

Men ging dien dag aan het werk om den koers iets juister te kunnen stellen, kraste te
dien einde in het hout van de plecht een kaart van de eilanden Sumatra en Java en Straat
Soenda,—alles op het geheugen. Den middag vóór het ongeluk had Bontekoe vijf-en-een-
halven graad Zuiderbreedte bevonden; het bestek op de kaart wees toen negentig mijlen
tot de kust. Van dat punt uit stelde men zijn koers. Drommels, had men maar een
kwadrant!

„Heeft niemand een passer?” vroeg Bontekoe.

„Daar vraag je zoowat, schipper”, antwoordde Teunis Sijbrandt, de kistenmaker. En hij


diepte uit zijn broekzak een passer op. „’t Is boffen: meest legt ie op m’n tafel!”

„Geef hier!” zei Bontekoe verheugd. „Dan zullen we eerst eens ’n gradenboog snijden!”

Zoo gebeurde. Men trok in een plankje een zoo groot mogelijken kwartcirkel, mat daarin
alle graden uit en bracht daarna den wijzer aan. Den volgenden middag nam men er, zoo
goed en zoo kwaad als het ging, hoogte mee en stelde den koers op Sumatra.

Zoo zeilde men verder, overdag koers en hoogte nemend op de zon, en ’s nachts op de
sterren. Den derden dag was het brood op. Den dag te voren had ook de dorst zich reeds
geducht doen gevoelen. Maar men bleef vol hoop. De wind zat achter in ’t zeil; de zee
bleef kalm.

Den dag daarop doemden zwarte wolkgevaarten aan den gezichtseinder op. Geweldig
schreden ze nader, den ganschen hemel als pad nemend. De maats kenden die wolken!
In opgewonden vreugde werden de zeilen horizontaal gespannen. Pik-zwart was nu het
gewelf; het scheen den mannen, dat ze in een grooten, donkeren kelder zaten. De zee,
die dagenlang een felblauwen hemel teruggekaatst had, slurpte al die zwartheid gretig
op,—leek wel een modderpoel. [244]

Daar kletterde de regen neer.

De levenskracht ontwaakte weer; de harten zwollen. In een oogwenk was het zeil vol.
Men kon nu de twee vaatjes vullen, die als bergplaats voor het brood hadden gediend.

Een koude nacht volgde. De maats bibberden in hun doorweekte kleeren.

Maar den volgenden morgen schroeide de zon ze in een ommezien droog en deed de huid
vervellen.

Snikheet werd het. De zee was glad als een spiegel. Waar zich te bergen voor de
gloeiende zon? De dorst deed zich weer gevoelen. Bontekoe sneed de neuzen van zijn
schoenen af en liet allen een „beker” vol uit de vaatjes geven. Daarmee was drie-vierde
van den voorraad op, want men moest deelen met de makkers in de sloep, die niets
hadden om het water te bergen.

Vreeselijke dagen volgden. Als een verschrompeld stukje leer zat de tong in den mond;
keel en verhemelte schroeiden; het ontberen van voedsel bracht krampen in de
ingewanden teweeg. Telkens wanneer de morgen grauwde, hoopte men land te
ontwaren. Telkens weer nieuwe teleurstelling, wanneer men niets dan zee zag, zoover het
oog reikte.

Harmen vond het noodig, den kerels wat moed in te blazen. Bij gebrek aan z’n fiedel, die
in de vlammen een einde had gevonden, kwam hij met een van zijn gewaagde „verhalen”
op de proppen.

„Ik zal jullie vertellen, hoe het m’n oom gegaan is, jongens! Die heeft een tapperij voor
zeelui,—voor landrotten tapt ie niet. Z’n leven lang heeft-ie gevaren, van z’n tweede tot
z’n acht-en-zestigste. Zeven-en-twintig reizen heeft-ie gemaakt, waarvan drie-en-dertig
met schipbreuk! En overal goed afgekomen behalve een krab op z’n wang en da’s van het
baardscheren. Hij zat eens vast op een rif in de Chineesche zee, waar de visschen en de
menschen staarten an d’rlui knikker hebben. Goed, d’r komt een storm, de schuit vliegt
aan flarden, z’n zes-en-veertigste schipbreuk; m’n oom en de bottelier zijn de eenigsten,
die in de sloep komen. Eten aan boord? Geen spiering! Hongerlijje maar! Toen ze zeven-
en-tachtig dagen niks gegeten hadden, zei de bottelier: „’k Zou wel een hapje lusten!”—
„En ik”, zei m’n oom. „’k Zou jou wel lusten”, zei de bottelier. [245]„Ik jou ook wel”, zei m’n
oom. Goed, ze krijgen verschil van meening. „Als jij mij jouw beenen geeft, zal ik jou m’n
Zondagsche pet geven”, zei de bottelier. „Jawel”, zei m’n oom, „je kunt er een por met m’n
mes bij krijgen.”—„In je mes hap ik niet”, zei de bottelier. „Nou, laten we d’r dan om
loten!” zei m’n oom. Ze smijten de steenen. Allebei acht. Nog er eis! Wéér gelijk! En
weer!! Na dertien-en-een-halve dag zei m’n oom: „Vooruit, ik heb geen aardigheid meer
aan speulevaren. Snij me maar aan mootjes! Groet m’n wijf en neem m’n gouwe ring
maar voor de moeite.” Temet, dat ie zich omkeert om niet te zien, hoe de-n-ander ’m zal
afmaken, ziet ie, dat de sloep.…. op het strand zit! „Land!!” roept ie. Hadden ze me daar
al elf dagen aan land gelegen en er door al dat dobbelen niks van gemerkt! Wat zeg je
me daarvan?”

„Mooi!” zeiden de oomes.

Toen viel het zwijgen weer in.

[246]
[Inhoud]
HAAIEN
Zonder dat iemand iets merkte, streek een gast tusschen de
schipbreukelingen neer. Eerst toen hij er was, voelde men zijn
aanwezigheid. De mannen hoorden hem in hun eigen matte, schorre stem
en zagen hem in elkaars fletse oogen.

Wanhoop heette de gast.

Wanneer een maat iets meende te ontdekken, wat land zou kunnen zijn,
greep men in koortsige haast de riemen en trok ze hijgend door het water.
Dan week de gast, geruischloos als hij gekomen was. Maar als even later
het „land” zich weer in lucht en water oploste, kwam de indringer terug. De
mannen ontweken elkaars blik, om hem maar niet te zien, en zij zwegen,
om door hun stem elkander niet te verraden, dat hij er weer was.
Beklemmend was het zwijgen; het snoerde de ziel toe. Als er iemand
kuchte, schrokken de anderen even en spitsten het oor, of er wat volgen
zou.

„Wat wou je zeggen?” vroeg een oome.

„Ik? Niets. Waarom?”

„Wel, je kuchte, en toen dacht ik: hij wil zeker wat zeggen.”

„Neen, ik hoestte zoo maar.”

„Nou, ik dacht het ook alleen maar.”

Dan viel het zwijgen weer in.

---------------------

Op een middag groote opschudding. Uit het Oosten kwamen meeuwen


aanvliegen, wel dertig, die krijschend om de booten cirkelden en er nu en
dan zoo laag overheen streken, dat [247]men ze bijkans grijpen kon. In de
sloep lag een roestige degen,—daarmee ging Hilke op de plecht staan, en
onder heesch gebrul wist hij er een mee vleugellam te slaan. En zoo had
men er, toen het duister was, vijf weten te vangen. De vogels werden
geplukt en verdeeld. Met van begeerte trillende handen namen de oomes
het luttel brokske vleesch, dat hun was toegedacht. Ze kauwden er over,
zoolang het maar ging, en zogen halsstarrig aan de merglooze
vogelbeentjes. In de hoop, dat ze morgen de andere vogels ook nog buit
zouden maken, gingen de oomes den nacht in.

Maar toen het eerste licht schemerde, waren de meeuwen weg.

Toch was er hoop ontkiemd. En door de eerlijke verdeeling der kleine


vangst over sloep en jol was het gevoel van saamhoorigheid weer versterkt:
men besloot, ondanks het gevaar, dat er aan verbonden was, de makkers
uit de sloep nu toch maar in de jol op te nemen. Want de olie in de lantaren
was sinds lang opgebrand en daarmee de kans om ’s nachts uiteen te
geraken weer belangrijk vergroot. Men kon nu ook den mast en de zeilen
van de sloep overnemen en zeilde bijgevolg met een grooten mast, een fok,
een bezaan en een blind zeil. Toen tegen den avond de wind toenam,
merkte men tot algemeene vreugde, dat de jol, ondanks haar zwaardere
belasting, sneller voer.

Vreemd, men rekende nooit met de mogelijkheid van ’s middags of ’s


avonds land te krijgen. Dat werd alleen ’s morgens, bij zonsopgang
verwacht. Overdag scheen het, als schoot men in het geheel niet op: er
was niets, dat men naderbij zag komen of verdwijnen, en de horizon bleef
altijd gelijk. Slechts de wolken trokken voorbij, maar die kwamen van
achter en verdwenen weer ver vooruit, zoodat men het gevoel had van
terug te blijven.

Maar ’s nachts! Je hoorde het klotsende water, door den boeg op zij
geworpen; voelde den wind aan het zeil trekken, en achter je zag je in de
donkere watermassa de lange, lichte streep, die er op duidde, dat er
voortgang in de kast zat! Wie weet, of ze nou al niet land voor den boeg
hadden; wie weet, of ze morgenochtend niet vlak voor hun oogen de
boomen zouden zien oprijzen.…. Was dat het ruischen der branding al
niet?!

Dan kwam de langverbeide ochtendklaarte.—Zee. Niets dan zee. [248]

Sedert vijf dagen hadden de oomes geen druppel water meer over de
lippen gehad.

Achter de boot schaarde zich een afschuwwekkend gevolg. Toen een oome
den eersten witten haaienbuik in het water zag flitsen, slaakte hij een kreet
van schrik en walging. Een enkele maal stiet een maat met een huivering
den degen in het water, en allen rilden van lol, wanneer een roode
bloedschemering verraadde, dat de stoot doel had getroffen.

Het vreeselijkst van al deed de dorst zich gelden. Men kauwde op sleutels
en musketkogels, om het dorre verhemelte nog wat speeksel af te
scheiden.

Toen gebeurde weer iets, dat den moed herleven deed. Een school
vliegende visschen dook, waarschijnlijk uit vrees voor de haaien, vlak voor
de boot op. Maar met vieren, vijven tegelijk tuimelden ze tegen de zeilen en
vielen den ijverig grabbelenden oomes ten buit. Ze werden rauw
verslonden, smaakten fijner dan de fijnste zalm.

En men dobberde weer verder.

Den eersten December—den twaalfden dag, dat men in de booten was!—


begonnen enkele maats, ondanks Bontekoe’s en Vader Langjas’
waarschuwingen, zeewater te drinken. Daar het den dorst niet leschte,
zwolgen ze maar door, tot de maag haar weerzin ervoor te kennen gaf en
alles weer naar buiten wierp,—tot het geluk der oomes, die anders stellig
ziek zouden zijn geworden. Nu brandde hun keel meer dan ooit, en hun
dorst was nog toegenomen. De tranen rolden den armen kerels over de
wangen.

Floorke had zich een snee in den bovenarm gegeven en zoog zich het bloed
uit.
Vader Langjas, uit wiens oogen alle levenskracht geweken was, stelde voor,
de jol lek te stooten en zich met z’n allen te laten zinken.

„En dan de haaien tot voedsel dienen?” vroeg Bontekoe.

Daar had geen enkele oome trek in. En met nieuwe bezieling, vast besloten
om, zoolang er nog een vonkje leven in hun uitgehongerd karkas zat, het
niet als haaienvoedsel te laten dienen, keken de mannen weer uit, naar het
Oosten.….

De haaien waren geduldig,—verlieten de jol niet. [249]

Allengs zakte de moed weer. Een paar maats begonnen te schreien,—


wilden zich overboord werpen.

„Wat drommel”, zei Bontekoe driftig, met schorre stem, „als die stomme
dieren de moed niet opgeven, zullen wij het dan doen?”

Maar bij sommigen was het laatste restje levensmoed gebroken. Het zou
wel niet voor de eerste maal zijn, dat deze haaien een jol met
schipbreukelingen volgden; ze zouden wel weten wat ze deden!—Met holle,
koortsige oogen tuurden ze in het water, krompen ineen, wanneer daar
beneden iets donkers voorbij schoot.….

„Hajo”, kreunde Padde. „ik kan niet meer, Hajo! Ik wil liever dood gaan.”

Hajo zocht naar bemoedigende woorden. Maar woorden, dat bleven het.
Padde voelde de holheid ervan en verloor zijn laatste aasje moed, nu hij
merkte, dat ook zijn vriend de wanhoop nabij was.

Bontekoe ging het als Hajo. Hij moest zijn zeventig groote kinderen
troosten—en zocht zelf naar troost.….

Rolf zei niets, staarde uren lang naar den Oostelijken horizon. Het kwam op
volhouden aan. Volhouden.

Den volgenden dag regende het! In zenuwachtige haast, gepaard met


driftige, ruwe uitroepen, spanden de oomes het bezaanszeil en het blinde
zeil boven de boot, vulden de beide vaatjes weer, verzamelden het water
verder in schoenen, leeren mutsen, en slurpten gulzig uit het zeil. Toen
kropen allen weer in de holte van de jol bijeen om wat warmte te zoeken.
Hun kleeren waren doorweekt, en een vochtige morgenkilte hing boven het
water. De lucht was troosteloos grijs, en hoewel de maats nu weer tijdelijk
van hun vreeselijksten kwelgeest bevrijd waren, staarden ze onder het zeil
door met hun blauw-omrande oogen triest in den dikken regen-nevel, en
het doffe zwijgen der laatste week viel weer in.

Een schorre kreet.….

„Land!!”

Als versteend blijven allen zitten, de oogen wijd open, angst en twijfel in
het gelaat. Men durft haast niet opstaan en zich overtuigen. Wie zou nu
nog een teleurstelling kunnen dragen? [250]

Maar de man aan het roer is zeker van zijn zaak. „Land!! Land voor den
boeg!!” De tranen sidderen door zijn stem.

Dan krabbelen allen met hun verstijfde ledematen overeind, steken hun
koppen onder het zeil door en.….!! Daar, aan den Oostelijken
gezichtseinder.….!!! Enkelen gillen hun blijdschap uit, anderen staren
sprakeloos of stil schreiend naar het grijs-blauwe streepje in de grijs-blauwe
verte.

Met alle man zette men hijgend en vloekend de zeilen weer bij. Zoo werd
langzaam-aan het streepje land grooter. Men onderscheidde bergvormen,
de lichte streep der branding, daarachter groene bosschen. Sumatra kon
het niet zijn: het was een eilandje, waarvan men den geheelen omtrek
overzien kon. Maar Bontekoe wist, dat Westelijk van Sumatra, dicht op de
kust, een rij eilanden liggen. Daar moest dit er een van zijn.

Toen men het eilandje naderde, bleek de zee geducht aan te loopen.

„We moeten een landingsplaats zoeken!” zei Bontekoe.


„We kunnen hier toch landen, schipper? Je zult zien, dat alles goedgaat!
Nietwaar, jongens?!”

„Ja-zeker!” brulde de heele schaar.

Maar hier werd de schipper weer de schipper. „Zullen we nu op het laatste


oogenblik alles bederven? Zeg op: wie heeft jullie naar land gevoerd? Heeft
de Neus dat soms gedaan? Of jij, Floorke? Of Gerretje, of Hilke?”

„Neen, schipper, dat heb jij gedaan.”

„Geloof me dan ook, als ik jullie zeg, dat we hier nooit heelhuids door de
branding komen. We moeten een betere plek zoeken.”

Hij werd gehoorzaamd. Men voer het eiland om en vond aan de binnenzijde
een kreek. Men roeide ze in, liet de dreg vallen, klom, zoo goed en kwaad
de leden het nog veroorloofden, de boot uit, waadde door het ondiepe
water.

Schreiend kusten de arme kerels het strand.

[251]
[Inhoud]
JOPPIE III
Toen de mannen over hun eerste en hoogste uiting van blijdschap heen waren,
kropen enkelen het strand over naar het bosch.

Er groeiden overal kokosboomen, en de noten lagen voor het grijpen op den


grond. De meeste waren in den val gebroken; die kon men met de handen verder
splijten. En dan de tanden in het witte vleesch gezet.….! Toen de maats zooveel
vruchten naar binnen hadden gewerkt, dat er geen stukje meer in wilde,
verlangden allen naar rust, naar niets anders dan rust. Ze sleepten wat dor gras
en bladeren aan. En toen.….! Hè! Hoelang was het geleden, dat ze zich voor het
laatst behoorlijk hadden kunnen uitstrekken, en dan nog wel op zoo’n zacht leger
en zonder den hongerdood voor oogen! Nu zou verder alles ook wel goed gaan!
De schipper was bij hen en kende den weg! Met een oneindig dankbaar gevoel
sliepen allen in.

Maar weinige uren later werd de een na den ander met hevige buikkrampen
wakker. De ingewanden bleken niet tegen zooveel voedsel opeens bestand te zijn.
Allen kropen bijeen en klaagden over hun onverdragelijke pijnen.

„Gevert! Ik ga d’r an! Gevert! O, God, m’n buik.….”

„Was ik maar thuis, bij m’n wijf! Die wist er wel raad op.… Doris!”

Toen zakte de pijn weer. De mannen sliepen nog een paar uren en werden
eindelijk door de zon gewekt.

Ze stond boven de baai, en het hemelsche goud vloeide in het water neer. Rappe
strandloopertjes trippelden op hooge pootjes heen en weer; meeuwen streken
zwierig door het schuim der branding, of wiegden zich op het zacht deinende
water in de baai. Allerlei bontgepluimde vogels schetterden [252]en floten en
kwinkeleerden in de hooge, statige boomen; in de wuivende kronen der kokossen
voerden grijsbruine klapperratten hun gedurfde luchttoeren uit.

De mannen voelden hun krachten herleven. Ze lagen daar in het reeds warme
zand, boven zich de stralende zon, die spoedig al te heet zou worden. Zwijgend,
stil genietend, luisterden ze naar het concert van de branding.
Bontekoe liet een gezamelijk gebed houden. Men zong een paar psalmen. Zelden
hadden de oomes met een dieper gevoel van dankbaarheid gebeden dan op dien
wonderlijk schoonen morgen aan het strand van dat Sumatraansche kust-
eilandje.

Daarna ging men eens op verkenning uit. Een groep toog naar het Zuiden, het
strand langs, een andere groep naar het Noorden.

In den namiddag kwamen ze terug met niets dan kokosnoten, bananen en enkele
andere, onbekende vruchten. Menschen hadden ze niet gezien. Maar zij, die naar
het Noorden waren getogen, hadden een grappig avontuur gehad.

Ze hadden een vlerk-prauwtje gevonden, waaruit ze hadden opgemaakt, dat het


eilandje bewoond moest zijn, al zagen ze dan ook geen sterveling. Ze hadden het
bootje, dat slechts twee man bergen kon, eens in zee geduwd, maar toen ze er
voor de grap met z’n drieën in waren gaan zitten, hadden de half vergane
planken zich begeven, en de spelevarende gasten waren kopje-onder gegaan.
Wat met dit zonnetje niet zoo slim was.

Bij de plaats, waar het prauwtje lag, ging een paadje het woud in. Men liep het
binnen, in ganzenmarsch, omdat er voor twee naast elkaar geen plaats was. Hilke
voorop!

Nauwelijks hadden ze het pad ingeslagen, of voor hen uit, vlak bij, begon een
hond te janken. In alle omzichtigheid werd nu de tocht voortgezet. Spoedig zag
men tusschen de dikke bamboestelen een open plek schemeren en in het midden
daarvan een bouwvallig huisje op hooge palen, aan een waarvan een magere
hond was vastgebonden, die allerafgrijselijkst te keer ging. Overigens zag het
huisje er nogal vreedzaam uit: een paar duiven vlogen van het dak op, en toen
kwam zelfs een duif uit het lage huisdeurtje fladderen, zoodat het heele
gebouwtje meer op een til dan op een menschelijke woning geleek. Weifelend
betraden de maats de open plek. [253]

Toen de hond zijn gasten zag opdagen, staakte hij zijn Jobsiaansche jeremiades,
jankte vreugdevol, ging van opwinding op zijn achterpooten staan en verhing zich
daardoor bijkans in den strik om zijn hals. En toen de oomes hem den smerigen,
stekeligen kop krauwden, kreunde hij zacht van geluk, kwispelstaartte en draaide
dankbaar het achterlijf.
Een gekorven kokos-stam stond schuin tegen het huisje op,
—scheen als „trap” te hebben dienst gedaan. „’t Lijkt wel
een kipperen”, meende Floorke, terwijl hij naar boven
balanceerde. „Blijf jij nou beneden!”—dat was tegen
Gerretje, die volgen wou—„We kunnen er met z’n tweeën
niet op: ’t is geen marremeren trap!” „Wat zie je?” vroegen
ze van beneden, toen Floorke naar binnenkroop.

„Pah!” gromde Floorke en spuwde luidruchtig. „’k Heb een


spin in m’n mond!”

„Kun je wat zien, of is het donker?”

„Donker? ’t Dak is zoo lek als de Zuid-Westenwind! Maar d’r


is niks te zien! ’n Paar gebarsten potten! Wacht, daarachter
is nog een kamertje, geloof ik.”

„Ga daar ’ns kijken?” En toen Floorke geen haast maakte,


smaalden ze: „Of durref je niet?”

„’t Is ijs van één nacht, die vloer”, aarzelde Floorke. „Je kijkt
d’r zóó doorheen.”

„Wat een vent!” hoonde Gerretje van omlaag. „Kom terug,


dan zal i k gaan kijken!”
„’t Lijkt wel een
kipperen!” meende
„Als je d’r trek in hebt, vooruit maar”, zei Floorke en
Floorke.
klauterde naar beneden.

Gerretje’s gelaat drukte thans twijfel uit. Hij wist wel, dat Floorke niet voor een
klein geruchtje vervaard was. Als die zei: „’t Is vuil!” dan was ’t ook vuil. Maar
eens gezegd bleef gezegd. Hij klauterde behendig de „loopplank” op, zooals hij
zei, en overzag, boven aangekomen, het terrein. „Dunnetjes is ’t!” moest hij
toegeven.

„Oh!” stelde Floorke beneden vast.

„Weet je wat ik doe?” zei Gerritje. „Ik spring er overheen! Daar ginds is wéér een
dikke bamboe!” [254]

„Ja, laat je niet kisten, Gerretje!” riepen de maats. „Je botten zijn betaald!”
Gerretje zette af, van een-twee-dr.….! Een heidensch gekraak. De „trap” sloeg
neer, kletste in de modder, die alle zijden uitspatte, en het huisje viel keurig
netjes om, de vier ontwortelde palen in de lucht, zoodat de hond, die aan een der
palen gebonden was, met uitpuilende oogen in de lucht kwam te bengelen. En
het dak scheurde open, en daaruit buitelde, als een Sinterklaasverrassing.….
Gerretje.

De maats zakten bijkans ineen van plezier. Floorke sneed haastig den hond los,
die daarop half flauw op den grond tuimelde en het als zijn eerste plicht
beschouwde, Floorke’s bloote voeten schoon te likken. Gerretje krabbelde met
een vrij krasse bewering tegen de Sumatraansche huizenbouw overeind en
veegde de handen aan zijn broek af.

„Laten we er het zwijgen maar toe doen!” stelde Floorke voor. En terwijl er een
paar in het omgevallen huisje snuffelden en veel stof, spinnen en kakkerlakken
vonden, liepen de anderen de open plek eens rond en ontdekten een smal pad,
echter zóó verwilderd, dat men zich slechts met den bijl een doortocht zou
kunnen banen.

Men besloot terug te keeren en nam den hond als krijgsbuit mee. Uitgelaten van
vreugde sprong het beest tegen zijn bevrijders op, die intusschen de meest
fantastische veronderstellingen te berde brachten over den samenhang van de
vergane prauw, het wrakke huisje en den uitgemergelden hond.

Zoo kwam men met een levende ziel méér bij de jol aan. De mannen, die onder
leiding van Folkert Berentsz. waren uitgegaan, vertelden, eenigszins jaloersch op
den vierpootigen buit der anderen, een slang te zijn tegengekomen, zoo dik, dat
ze hem met z’n vijven niet omspannen konden. Hij had gesist, dat je er koud van
werd. Padde wou hem bij z’n staart pakken—nietwaar, Padde?—toen ie er van
tusschen ging, de boomen wegdrukkend als grashalmen.

Maar toen wist Floorke te vertellen van een krokodil, wiens staart ze uit het bosch
hadden zien steken, en toen ze het eiland half waren omgeloopen, hadden ze aan
den anderen kant zijn kop gevonden, een merakel klein koppie, niet grooter dan
de N i e u w - H o o r n .…. destijds. Gerretje had z’n kop voor [255]een kloof
aangezien, want het toeval wilde, dat de krokodil juist gaapte. Gerretje was er in
geloopen en had nog gezeid: „Jongens, wat is de grond hier slappies!”—Die
grond was natuurlijk de tong van die krokodil geweest. Nou, en in eens had het
mormel z’n bek dichtgeslagen, en Gerretje zat in het pikkedonker. Tusschen de
tanden was hij er weer uitgekropen, waar Gerretje?
Maar toen had Harmen nog heel wat anders te vertellen! Hij had er over willen
zwijgen, maar nu Floorke zóó begon, zou Harmen zijn broek ook eens in ’t
zonnetje hangen! Nou, ze hadden me dan een beestje gevonden, zoo op het
eerste gezicht een regenwurm. De Schele had het bij zich gestoken voor Vader
Langjas, maar ineens was het in zijn zak begonnen te praten. „Schele”, had het
gezegd, „scháám jij je niet?” Toen had de Schele het beest weggegooid; het had
in zuiver Maleisch: trima kassi banjak! gezeid en was toen in zee
weggezwommen,—had, met een blad boven z’n kop als blind zeil, tegen den wind
in gelaveerd.

Tegen zulke avonturen voelde Floorke zich niet meer opgewassen. „Laat je duim
eens kijken?” zei hij smalend tot Harmen. „Je zult ’m wel heelemaal plat gezogen
hebben.”

Harmen grinnikte, toonde z’n duim.

Men laadde de jol met kokosnoten en bananen, borgde vruchten onder de plecht,
in het achterkastje, boven op het roefje … het leek wel, of de jol ter markte toog.
Een riviertje was niet gevonden, zoodat men de vaatjes maar met kokosmelk
vulde.

Den tijd, dien de mannen bezigden om te proviandeeren, maakte de magere,


stekelige hond zich ten nutte door een groote boschrat te vangen en die in
gulzige haast te verorberen. Toen hij in de gaten kreeg, dat de mannen van plan
waren in zee te steken, vroeg hij op hondenmanier om meegenomen te worden.
Hij scheen met de zee vertrouwd te zijn.

De maats willigden zijn verzoek in. Ze klopten hem vertrouwelijk op z’n bottige
flanken, stelden vast, dat je z’n ribben tellen kon, dat z’n ooren allergemeenst
lang en steil waren, dat z’n vel bij den vilder geen duit zou opbrengen, maar dat
hij, naar z’n oogen te oordeelen, een rondborstige natuur had.

Hij werd ook gedoopt.

„Joppie” noemden ze hem—want driemaal is scheepsrecht. [256]

In de schemering verliet de jol het land. Verder maar weer! De maats waren vol
moed. Hier konden ze toch niet blijven, en de schipper zei, dat ze binnen twee
dagen Sumatra voor den boeg zouden krijgen. Dus nog maar eens het lijf
gewaagd! Het gevoel van onrust, dat hen overmeesterde, toen ze het eilandje in
de duisternis zagen wegzinken, werd dapper weggeslikt.
Joppie hief als afscheidsgroet een erbarmelijk gehuil aan, dat uit de verte door
apengekrijsch beantwoord werd. De oomes stelden hem voor de keus: over
boord te vliegen of met z’n gegil op te houden. Joppie verkoos het eerste, werd
door Hilke bij z’n nekvel ferm in het water ondergedompeld, schudde toen z’n
natte, steile haren uit, zocht daarop achter in de roef het beste plekje op en
sluimerde zuchtend in, tot Folkert Berentsz. hem deed verhuizen, omdat hij daar
liggen wou. Toen namen een paar oomes Joppie als hoofdkussen,—waarvan ze
later geduchte spijt en kriebel kregen.

De maan kwam op, mild, vriendelijk als een moederoog, wakend over de zeventig
brave jongens in de jol.

Een groote zwerm kalongs streek hoog over de jol in Oostelijke richting voorbij,
zwijgend,—als schaduwen.

Alom stilte. Stilte.

You might also like