authors@Fred Hutch

Frost, Sophia

June 15, 2015 SCIENCE SPOTLIGHT 
 
1 Volume 5, Issue 6 | Fred Hutchinson Cancer Research Center 
 
Timing of MDS Relapse: A Multifaceted Case to 
Crack 
June 15, 2015 
    SHL Frost 
Myelodysplastic syndrome (MDS) comprises a group of hematological disorders in which 
hematopoietic stem/precursor cells in the bone marrow do not develop properly into healthy blood 
cells. Allogeneic hematopoietic cell transplantation (HCT) represents a possible cure, but 10 to 50 % 
of patients relapse, which is suggestive of a poor outcome. Most relapses occur within 12 to 18 
months after HCT. However, there is no defined time point beyond which the patient is considered 
"safe" - some relapses are diagnosed several years after transplantation. 
The question thus arises: what factors determine who will relapse and when? This occupied the 
Clinical Research Division's Drs. Cecilia Yeung and Joachim Deeg, whose results from analysis of 
relapse patterns among 1,007 transplanted MDS patients were recently presented inBiology of 
Blood and Marrow Transplantation. "This is the first large study to examine risk factors for and 
features of late relapse after HCT for MDS," Dr. Yeung said, further explaining that this kind of in-
depth look at late-relapse clonal characteristics could help identify patients who might benefit from 
targeted interventions after HCT to prevent disease progression. 
Disease progression and relapse after HCT are mainly determined by changes in the structure or 
function of cells and chromosomes, referred to as clonal cytogenetic abnormalities, likely supported 
by somatic (i.e. non-inherited) mutations. Although there is limited knowledge about the prognostic 
effect of molecular abnormalities, recent findings indicate that MDS clones present at the time of 
post-HCT relapse often differ from those identified pre-HCT, although the "relapse clones" have yet 
to be properly characterized. 
In the patient cohort analyzed by Dr. Yeung and colleagues, relapses before 18 months were 
categorized as "early" and those after 18 months as "late", and differences in risk factors between 
the two data sets were assessed through logistic regression analysis. From the initial cohort, 973 
MDS patients were included; of those, 25 % eventually relapsed; 213 before 18 months (median 100 
days), and 41 after 18 months (median 2.6 years). As expected, the hazard of relapse in the entire 
cohort declined steadily, though without any distinct time point at which the trend changed 
significantly. The difference in median survival after relapse was not statistically significant between 
early- and late-relapsing patients (91 and 220 days, respectively; p = 0.07). Multivariable regression 
June 15, 2015 SCIENCE SPOTLIGHT 
 
2 Volume 5, Issue 6 | Fred Hutchinson Cancer Research Center 
 
analysis of patients who were relapse-free and alive at 18 months (n = 408) identified a number of 
risk factors, including the finding that MDS that had developed into acute myeloid leukemia implied 
higher risk of late relapse than untransformed MDS. Another conclusion was that late relapse was 
less likely among patients with chronic graft-versus-host disease (GVHD) compared to those without 
chronic GVHD - a finding that concurred with the documented graft-versus-leukemia/tumor effect. 
"The alloreactivity of donor cells is a major contributor to the eradication of the patient's disease," 
explained Dr. Yeung. 
In general, few significant differences were noted in characteristics of early- and late-relapsing 
patients. Poor-risk cytogenetics were associated with higher probability of early relapse, as was pre-
HCT conditioning with reduced-intensity regimens. Genetic analysis performed on sequential bone 
marrow samples from 36 late-relapsing patients revealed additional clonal abnormalities at relapse 
besides those detected in pre-HCT samples in 41 %; 30 % had their pre-HCT abnormalities replaced 
by new clones; 17 % exhibited the same clone pre-HCT and at relapse; and in 10 % no 
abnormalities were revealed for either time point. 
Three patients in the "late" group were further analyzed by chromosomal genomic array testing of 
pre-HCT and post-relapse DNA, to examine potentially prognostic alterations that may be 
undetectable by standard methods. Interestingly, post-relapse-unique abnormalities were discovered 
for all three patients. Still, higher-sensitivity techniques may be needed to confidently exclude the 
possibility of the deviations being present at very low levels already before HCT. 
Dr. Yeung concluded that further refined mutational analyses should be undertaken. "We 
hypothesized that late relapses may be due to an 'escape' from surveillance (by donor alloreactivity) 
due to the evolution of new clones/subclones. However, results do not show clear differences in 
clonal architecture between early and late relapse." The team, in collaboration with Dr. Jerry 
Radich's laboratory, will now continue the studies of MDS progression dynamics and molecular 
profiles through single cell techniques and massive deep sequencing, in addition to exploring 
emerging questions regarding late relapse versus "new disease", which might have to be handled 
differently. 
Yeung C, Gerds AT, Fang M, Scott BL, Flowers ME, Gooley T, Deeg HJ. 2015. Relapse after 
allogeneic hematopoietic cell transplantation for myelodysplastic syndromes - analysis of late 
relapse using comparative karyotype and chromosome genome array testing. Biol Blood Marrow 
Transplant [Epub ahead of print]. 
 
June 15, 2015 SCIENCE SPOTLIGHT 
 
3 Volume 5, Issue 6 | Fred Hutchinson Cancer Research Center 
 
 
 
Image provided by Dr. Cecilia Yeung 
(a) Bone marrow karyotypes before HCT (broken pie chart) and at 
the time of relapse (intact pie charts) in 213 patients with early 
relapse. (b) Bone marrow karyotypes before HCT (broken pie chart) 
and at the time of relapse (intact pie charts) in 41 patients with late 
relapse. 
 


English

Timing of MDS Relapse: A Multifaceted Case to Crack

Berg, Barbara L

Neiman, Paul E

Groudine, Mark T

Fred Hutchinson Cancer Research Center History Project.

Pattacini, Laura

July 20, 2015 SCIENCE SPOTLIGHT 
 
1 Volume 5, Issue 7 | Fred Hutchinson Cancer Research Center 
 
A COMPASS to Navigate Oceanic Datasets 
July 20, 2015 
    L Pattacini 
 
 
Overview of an ICS experiment and 
COMPASS analysis. COMPASS 
outputs heatmap of posterior 
probabilities that the corresponding 
cell-subset (column) exhibits an 
antigen-specific response in the 
corresponding subject (row). 
 
Image provided by Dr. Lynn Lin. 
 
 
T cell responses that are generated in vivo by antigen exposure can be amplified ex vivo upon 
incubation of T-cells with the same antigen. These antigen-specific T cell responses can vary 
drastically in their quantity and quality, measured by the combinations of cytokines that are 
produced. These variations depend on the antigen as well as on an individual’s characteristics. Such 
variations make the study of the human immune system fascinating and challenging at the same 
time. A large number of possible combinations of cytokines produced by antigen-specific T cells 
complicate the analysis of large data sets generated to evaluate immune responses. As scientists, 
we should not discriminate between different subsets based on our preference but we should 
carefully look at all of them. 
Customized tools are needed to do so in the most unbiased and reproducible way possible. 
Although a number of tools are available for flow cytometry data analysis, very few are designed 
specifically for high-throughput analyses. Most importantly, none of them is dedicated to characterize 
the quality of the cytokine response. Things are about to change. In fact, in a study published last 
month in Nature Biotechnology, Dr. Lynn Lin from the Vaccine and Infectious Disease Division 
(VIDD) at Fred Hutch reported the results obtained by analyzing three sets of data using a new tool 
July 20, 2015 SCIENCE SPOTLIGHT 
 
2 Volume 5, Issue 7 | Fred Hutchinson Cancer Research Center 
 
named COMPASS. COMPASS allows running qualitative analyses of cytokine combinations 
produced upon in vitro stimulation by specific subsets of cells.  
The first data set was obtained from the RV144, a phase III clinical trial testing safety and efficacy of 
an HIV vaccine, which demonstrated a 31% vaccine efficacy. This dataset provided two benefits: 
first, it was possible to compare the results with previous analyses of the same data, and second, to 
correlate the results with trial outcome. The results from COMPASS not only showed increased 
sensitivity and specificity in discriminating between placebo and vaccine recipients as compared to 
two previously used methods, but also led to the identification of two polyfunctional cell subsets 
whose presence correlated with HIV protection in vaccinees. 
The second data set compared cytokine response to TB-specific peptides in tuberculosis (TB)-
infected versus uninfected subjects. Again, the use of COMPASS allowed for a better identification 
of the TB infected individuals. Furthermore, it identified antigen-specific subsets of cells in TB-
infected subjects that secreted cytokines other than IFN-gamma, which is the cytokine measured for 
TB diagnosis. The reported results suggest measuring more cytokines can lead to more sensitive 
assays to identify TB infection. 
The last data set analysis consisted of the comparison of two groups of individuals receiving the 
same vaccination regime, but at different doses. While the primary study showed no dose-specific 
differences, the use of COMPASS detected a reduction in one of the cell subsets when a higher 
dose of immunogen was used. 
Collectively, the three analyses presented showed that using COMPASS to examine not just the 
quantity but also the quality of antigen-specific T-cell responses can increase specificity and 
sensitivity and lead to novel discoveries missed by traditional approaches. In times of high 
throughput data analyses, COMPASS can point us in the right direction! 
Lin L, Finak G, Ushey K, Seshadri C, Hawn TR, Frahm N, Scriba TJ, Mahomed H, Hanekom W, Bart 
P-A, Pantaleo G, Tomaras GD, Rerks-Ngarm S, Kaewkungwal J, Nitayaphan S, Pitisuttithum P, 
Michael NL, Kim JH, Robb ML, O'Connell RJ, Karasavvas N, Gilbert P, C De Rosa S, McElrath MJ, 
Gottardo R. 2015. COMPASS identifies T-cell subsets correlated with clinical outcomes. Nat 
Biotechnol. DOI: 10.1038/nbt.3187. Epub ahead of print. 
 


A COMPASS to Navigate Oceanic Datasets

July 20, 2015 SCIENCE SPOTLIGHT 
 
1 Volume 5, Issue 7 | Fred Hutchinson Cancer Research Center 
 
Replacing the Western: an Improved Way to 
Measure Cell Signaling 
July 20, 2015 
     SHL Frost 
 
 
 
Immunoaffinity enrichment of a target 
peptide using the immuno-MRM assay 
developed in the Paulovich Lab. 
 
Illustration by Brad Singley 
 
 
 
 
Dysfunctional cell signaling is a common attribute in many disorders, and there is great interest in 
understanding the complex pathways that govern interactions and responses within our cells. These 
mechanisms can be studied through quantification of protein levels as well as covalent modification 
to proteins, so called post-translational modifications (PTM). A PTM with both biological and clinical 
relevance is phosphorylation - the addition of a phosphate group, which alters the function and 
activity of the target protein. Studies of phosphorylation signaling could help identify potential new 
drug targets, but a lack of adequate tools has stood in the way of progress - a situation that may be 
about to change. In a recent publication in Molecular & Cellular Proteomics, Drs. Jeffrey Whiteaker 
and Amanda Paulovich from Fred Hutch's Clinical Research Division describe a new multiplex assay 
that could provide the answer to many a researcher's litanies. 
Countless scientific advances have been made courtesy of traditional analytical techniques like 
western blotting and the enzyme-linked immunosorbent assay (ELISA), but these one-analyte-at-a-
time assays fail to live up to the demands placed by advanced modern-day "omics" technologies 
July 20, 2015 SCIENCE SPOTLIGHT 
 
2 Volume 5, Issue 7 | Fred Hutchinson Cancer Research Center 
 
(e.g. genomics, proteomics). Higher throughput, improved specificity, and better reproducibility are 
needed - preferably at a lower cost. Advances in the field of mass spectrometry (MS) has led to 
expansion of targeted MS, enabling precise quantification of protein-specific peptide sequences, the 
most commonly utilized form being multiple reaction monitoring (MRM). MRM-based assays provide 
direct measurement of the analyte and are recognized for yielding reliable output that can be 
reproduced independent of laboratory, thanks to the use of stable, synthetic internal standards. 
Compared with other techniques, these assays are highly specific, easily multiplexable, and 
comparatively quick and cheap. 
The Fred Hutch investigators set out to combine MRM with antibody-based affinity-enrichment using 
immobilized antibodies on magnetic beads, an immuno-enrichment method previously optimized by 
the Paulovich Lab (Whiteaker et al. 2007). To demonstrate proof-of-concept, an immuno-MRM 
assay was tested for quantification of phospho-signaling in an important network, the DNA damage 
response (DDR) network, which has been shown to commonly display mutations in tumors. 
Successful monitoring of variations in DDR activity could have major effects on basic, clinical, and 
public health studies. The resulting multiplex assay proved applicable for a wide range of samples 
and conditions, efficiently replacing no less than 69 western blots with one single 40-minute run. 
Dr. Whiteaker explained that the results were superior compared with those of semi-quantitative, 
often poorly specific, traditional methods, and that this moderate- to high-throughput MRM assay 
proved to be just what they hoped for: quantitative, precise, and specific. "Reliable methods for 
quantifying proteins and post-translational modifications are necessary tools for understanding the 
molecular basis of cellular signaling in diseased cells and profiling pharmacodynamic responses of 
new and existing therapeutics," he continued, underlining the broad impact of the innovation. 
From here, there are multiple possibilities to further advance the technique. "We are very excited to 
continue developing these assays, expanding assay content, and getting this technology into the 
hands of more biologists," Dr. Whiteaker concluded. There should be quite a few researchers out 
there that share his gusto. 
Whiteaker JR, Zhao L, Yan P, Ivey RG, Voytovich UJ, Moore HD, Lin C, Paulovich AG. 2015. 
Peptide immunoaffinity enrichment and targeted mass spectrometry enables multiplex, quantitative 
pharmacodynamic studies of phospho-signaling. Molecular & Cellular Proteomics. DOI: 
10.1074/mcp.O115.050351. [Epub ahead of print] 
July 20, 2015 SCIENCE SPOTLIGHT 
 
3 Volume 5, Issue 7 | Fred Hutchinson Cancer Research Center 
 
See also: Whiteaker JR, Zhao L, Zhang HY, Feng LC, Piening BD, Anderson L, Paulovich AG. 2007. 
Antibody-based enrichment of peptides on magnetic beads for mass-spectrometry-based 
quantification of serum biomarkers. Anal Biochem. 362(1):44-54. 
 
 


Replacing the Western: an Improved Way to Measure Cell Signaling

Zentner, Gabriel E.

September 21, 2015 SCIENCE SPOTLIGHT 
 
1 Volume 5, Issue 9 | Fred Hutchinson Cancer Research Center 
 
Picky proteins: amino acid preferences are all in 
the family 
September 21, 2015 
     GE Zentner 
  
 
 
 
 
 
 
Visual representation of amino acid 
preferences at selected sites in the 
H1N1 NP (left) and H3N2 NP (right). 
The height of a letter indicates the 
degree of preference for that amino 
acid. 
 
Image from the publication 
 
 
 
 
 
Evolution results in changes in protein sequence over time. However, these changes are not 
random. It has been appreciated for several decades that different sites within a protein evolve 
under different constraints, such that some sites can tolerate substitution to almost any amino acid 
while others may only be changed to a few amino acids. These constraints generally arise from 
interactions within a protein that shape its folding, stability, and function. An important question 
regarding site-specific amino acid preferences is how conserved they are during evolution. To 
address this, graduate student Michael Doud and postdoctoral fellow Dr. Orr Ashenberg in the 
laboratory of Dr. Jesse Bloom (Basic Sciences Division) performed a comprehensive 
characterization of the tolerance of amino acid positions to substitutions in two closely related 
influenza nucleoprotein (NP) homologs. They found that site-specific amino acid preferences are 
September 21, 2015 SCIENCE SPOTLIGHT 
 
2 Volume 5, Issue 9 | Fred Hutchinson Cancer Research Center 
 
highly conserved between the two proteins, enabling improved modeling of the evolution of other 
related proteins. 
The authors chose to study two NPs from two human influenza strains, H1N1 and H3N2, which are 
separated by 30 years of evolution and are 94% identical in sequence. To comprehensively assess 
the site-specific amino acid preferences of each site in these two NPs, the authors used deep 
mutational scanning. This is a technique in which PCR is used to create a library of plasmids 
encoding single-codon mutants, which are then incorporated into influenza viruses. Cells are then 
infected and grown to select functional NP proteins, and high-throughput sequencing is used to 
quantify the occurrence of each mutation before and after infection, the principle being that more fit 
NP variants would be overrepresented, while less fit variants would be underrepresented. 
Using deep mutational scanning, the authors were able to mutagenize 497/498 sites in each NP (the 
N-terminal methionine was not mutagenized) to all 19 other amino acids and measure amino-acid 
preferences at each site. Using these data, they then correlated the preferences between homologs. 
Strikingly, the correlation between the homologs’ preferences was nearly as robust as the correlation 
between preferences for a single homolog measured in this and a previous study. Analysis of 
individual sites revealed that, by and large, shifts in site-specific amino acids preferences were 
relatively small, with a consistently preferred amino acid across multiple biological replicates. 
Preferred amino acids were often conserved between homologs, but a small number of cases where 
a strong amino acid preference was different at a given site were observed. Further analysis 
revealed that some of these sites were evolutionarily variable, and that others were in functionally 
important regions of the protein, including its RNA-binding sites. 
The authors next used their experimentally determined site-specific amino acid preferences to inform 
models of NP substitution and trace the phylogeny of human influenza. Models trained with the deep 
mutational scanning data, particularly a combined dataset encompassing data for both homologs, 
performed much better than a non-site-specific model. Experimentally informed models also 
performed well in tracing influenza virus evolution in swine and equine hosts, indicating that site-
specific amino acid preferences are well-conserved across influenza NP from more distantly related 
hosts. 
This study shows that site-specific amino acid preferences are largely conserved across closely 
related protein homologs, which has important implications for extrapolating experimental data from 
one homolog to another. "Our study is important because in biology, we often perform studies of a 
specific gene but would like to extrapolate our conclusions to closely related homologs of that gene 
in other species. We therefore wanted to take advantage of new experimental techniques to make 
September 21, 2015 SCIENCE SPOTLIGHT 
 
3 Volume 5, Issue 9 | Fred Hutchinson Cancer Research Center 
 
the first measurement of how similar (or different) the effects of all possible mutations are to 
homologs of the same gene," said Dr. Bloom. "Our study provides a quantitative measure of how 
much one would expect the effects of mutations to change as a sequence diverges, and is therefore 
of great value in determining how to best formulate quantitative models of protein evolution." 
Doud MB, Ashenberg O, Bloom JD. 2015. Site-specific amino-acid preferences are mostly 
conserved in two closely related protein homologs. Mol Biol Evol [Epub ahead of print] 
Funding source: This work was supported by the National Institute of General Medical Sciences of 
the National Institute of Health (grant number R01 GM102198). M.B.D. was supported by NIH 
Training Grant T32 AI083203 and a fellowship from the Seattle Chapter of the Achievement 
Rewards for College Scientists Foundation. O.A. was supported by a PhRMA Postdoctoral 
Fellowship in Informatics. 
 
 
 


Picky proteins: amino acid preferences are all in the family

Koch, Lori

November 15, 2015 SCIENCE SPOTLIGHT 
 
1 Volume 5, Issue 11 | Fred Hutchinson Cancer Research Center 
 
Transcription factor binding sites in space and 
time: ChEC-seq 
November 15, 2015 
     L Koch 
 
 
 
 
 
In ChEC-seq, a chromatin protein of 
interest is genetically fused to 
micrococcal nuclease (MNase) and 
expressed in cells. Living cells are 
permeabilized and calcium is added to 
activate the MNase, leading to 
cleavage of DNA in proximity to binding 
sites for the protein. Total DNA is then 
purified and sequenced. 
 
Image provided by Dr. GE Zentner. 
 
 
 
 
 
 
 
Transcription factors bind to distinct locations in the genome and promote the expression of genes 
by recruiting RNA polymerase. Traditionally, transcription factor binding sites have been identified by 
chromatin immunoprecipitation (ChIP), which uses an antibody against the transcription factor to pull 
out the associated DNA for sequencing.  The DNA must be fragmented, either by digestion with 
micrococcal nuclease (MNase) or by sonication.  Thus, in order to keep the transcription factor 
bound to the DNA before the immunoprecipitation step, protein is chemically cross-linked to DNA 
before the digestion, usually using formaldehyde.  Formaldehyde preferentially generates protein-
protein crosslinks and so can repeatedly capture fortuitous interactions, such that they appear 
biologically meaningful.  Technically, ChIP can be challenging due to high background caused by the 
crosslinking, low quality antibodies, as well as poor solubility of the DNA binding protein.  Finally, 
traditional ChIP methods create a snapshot of the binding that happened during the crosslinking 
incubation, which is extremely useful but does not provide kinetic information about protein binding.   
November 15, 2015 SCIENCE SPOTLIGHT 
 
2 Volume 5, Issue 11 | Fred Hutchinson Cancer Research Center 
 
 
In order to remedy these issues, Dr. Gabe Zentner, previously in the laboratory of Dr. Steven 
Henikoff (Basic Sciences) and now at his own lab at Indiana University, developed a method that 
could characterize transcription factor binding sites on a genome-wide scale without crosslinking or 
an antibody and could also provide kinetic information.  Their method, termed ChEC-seq, was 
recently published in Nature Communications.  In chromatin endogenous cleavage (ChEC), a DNA-
binding protein of interest is fused to micrococcal nuclease (MNase), living cells are gently 
permeabilized, and then cleavage of the surrounding sequence, not protected by the protein, is 
induced by adding a high concentration of calcium.  Next, total DNA is extracted and fragments are 
selected by size for sequencing.  By using ChEC in combination with high-throughput sequencing 
(ChEC-seq), the authors were able to identify many more binding sites for the transcription factors 
they tested than had been reported by ChIP methods.  Furthermore, because the cleavage of the 
surrounding sequence by the MNase depends on calcium addition, the authors were able to 
determine which sites were bound and the surround sequence cleaved quickly or slowly, terming 
these two classes of binding sites "fast" and "slow." 
 
They wondered whether the differing binding kinetics of the two classes of sites reflected differences 
in the underlying sequence.  Indeed, their analysis showed that sites in the fast class had strong 
matches to the known consensus motifs of the tested transcription factors.  Conversely, the slow 
sites had lower motif scores. Importantly, the slow class of sites did not represent locations in the 
genome that were more open to transcription factor binding in general. In comparing datasets for 
each of the three major transcription factors tested, the authors found that the majority of the slow 
sites were unique to each transcription factor.  This suggests that the slow sites are preferred by a 
given protein and do not reflect accessibility biases.   
 
Recently, it has been suggested that the shape of the DNA can drive transcription factor binding in 
addition to sequence.  The authors wondered if the sites in the slow binding class were enriched for 
certain shapes of DNA and whether that could account for their binding more than whether they 
matched the consensus sequence.  They compared shape features of the DNA such as minor 
groove width, roll, propeller twist and helix twist between the fast and slow class of sites and found 
that the shape features were highly similar between the two. This result was consistent even when 
the authors ranked each site by motif match to account for similar sequences in each group. Finally, 
the authors compared the abilities of either a sequence or shape model to differentiate the two 
classes using L2-regularized linear regression. Their results strongly indicate that transcription 
factors recognize regular DNA shape patterns overall but that they only form long-lasting, stable 
interactions at sites with strong consensus motifs. 
November 15, 2015 SCIENCE SPOTLIGHT 
 
3 Volume 5, Issue 11 | Fred Hutchinson Cancer Research Center 
 
 
This work has illuminated key details about how transcription factors find their binding sites, which is 
an important part of understanding one of the first steps of normal and pathological biological 
processes.  Many genes are turned on in response to specific cellular conditions, how do 
transcription factors find the right targets so that the cell can respond? This has been a difficult 
question to answer without a way to distinguish binding over time on a genome-wide level.  
 
ChEC-seq has promise in distinguishing how other proteins interact with DNA over time.  Indeed, 
"ChEC-seq is a very simple method and, because it doesn’t rely on immunoprecipitation, gets 
around many of the issues associated with ChIP," said Dr. Zentner. "The major hurdle is generating 
an MNase fusion protein, but with the rapid adoption of CRISPR genome editing, I think it’s likely 
that generating MNase fusions for ChEC-seq in non-yeast systems will become routine." 
 
Zentner GE, Kasinathan S, Xin B, Rohs R, Henikoff S.  2015.  ChEC-seq kinetics discriminates 
transcription factor binding sites by DNA sequence and shape in vivo.  Nature Communications. 
6:8733. doi: 10.1038/ncomms9733. 
 
This work was funded by the National Institutes of Health and Howard Hughes Medical Institute.   
 
 


Transcription factor binding sites in space and time: ChEC-seq

December 20, 2015 SCIENCE SPOTLIGHT 
 
1 Volume 5, Issue 12 | Fred Hutchinson Cancer Research Center 
 
Congenital CMV prevention strategy: still 
missing 
December 20, 2015 
     L Pattacini 
 
 
 
 
CMV viral load in NFV-unexposed 
(left) versus exposed (right) infants. 
 
Image modified from the article. 
 
 
 
 
Although cytomegalovirus (CMV), or human herpesvirus 5, does not represent a threat for 
immunocompetent adults, congenital CMV infection is the most common cause of birth defects and 
occurs in 1 out of every 150 infants born (0.6%) in the US. The rate of CMV transmission is higher 
when the mother is HIV-infected, likely due to poor immune control of CMV replication. Although 
there are pharmacological treatments available for CMV infection, they have carcinogenic, 
teratogenic and gonadotoxic effects that preclude their use during pregnancy. Interestingly, nelfinavir 
(NFV), a protease inhibitor used for prevention of mother-to-child transmission of HIV (PMTCT), was 
shown to have an in vitro inhibitory effect on CMV replication by Dr. Soren Gantt, an Affiliate 
Investigator at the Vaccine and Infectious Disease Division at Fred Hutchinson. This discovery, 
together with the proven safety of NFV administration during pregnancy, led Dr. Soren Gantt, Dr 
Keith Jerome, and other investigators from US and Canadian research institutions to evaluate the 
effects of NFV treatment on prevention of mother-to-child transmission of CMV. The results of this 
study are published in the Journal of Medical Virology. 
The study analyzed specimens from children of 1,255 women enrolled in two prospective cohort 
studies coordinated by the International Maternal Pediatric Adolescent AIDS Clinical Trials 
(IMPAACT) Network. The studies were designed to assess the efficacy and safety of NFV and other 
December 20, 2015 SCIENCE SPOTLIGHT 
 
2 Volume 5, Issue 12 | Fred Hutchinson Cancer Research Center 
 
antiretroviral drugs for PMTCT. In both studies, NFV was prescribed to some women as part of 
clinical care and not part of a randomized regimen. A total of 314 infants were identified to have 
samples available and whose mothers received NFV for at least four weeks during pregnancy. 
Among them, 22% initiated NFV before pregnancy, 11% during the first trimester, 52% during the 
second and 15% during the third. Another 941 infants of women who did not receive NFV were 
included as controls. Mothers that did not receive NFV were more likely to know their HIV status 
before pregnancy and have a higher CD4+ count. All of the infants were HIV-uninfected. 
Approximately 2% of the infants had congenital CMV infection, a result comparable with previous 
studies during the era of combination antiretroviral therapy (cART). Infants with congenital CMV 
were born earlier and weighed less at birth than uninfected infants. 
NFV treatment was not shown to have a protective effect: comparable percentages of CMV-infected 
newborns were identified in the two groups (2.5% in the NFV versus 2.0% in the control group). 
Moreover, CMV viral load was not statistically different in the two groups (see Figure). This lack of 
association might be partially due to suboptimal timing of NFV use, a potential confounding factor in 
the analyses. However, mothers with a CD4+ T cell count lower than 200 cells/mm3 were more likely 
to have infants with CMV infection. Thus, although NFV was not found to have a direct effect on 
CMV infection, the results of the study provide further evidence that cART indirectly prevents 
congenital CMV infection by improving maternal immunity. According to Dr. Gantt, "this study 
underscores the importance of maintaining high CD4+ T cell counts to reduce congenital CMV 
infection in HIV-exposed infants, but also that unfortunately better preventive strategies are still 
needed." 
The study was supported by National Institutes of Health. 
Gantt S, Leister E, Jacobson DL, Boucoiran I, Huang M-L, Jerome KR, Jourdain G, Ngo-Giang-
Huong N, Burchett S, Frenkel L. 2015. Risk of congenital cytomegalovirus infection among HIV-
exposed uninfected infants is not decreased by maternal nelfinavir use during pregnancy. J Med 
Virol. doi: 10.1002/jmv.24420. [Epub ahead of print.] 
 
 


Congenital CMV prevention strategy: still missing

Neves, Alexandre

December 20, 2015 SCIENCE SPOTLIGHT 
 
1 Volume 5, Issue 12 | Fred Hutchinson Cancer Research Center 
 
The twists and turns in the life of Helicobacter 
pylori 
December 20, 2015 
     A Neves 
 
 
Smooth histograms summarizing speed 
distributions of bacteria in viscous 
solutions of gastric mucin. Top 
histogram: straight rod mutants show 
reduced swimming speeds (8-13%) in 
viscous solutions of gastric mucin. 
Bottom histogram: mutational 
perturbation of flagellum number shows 
a 19% increase in speed with 4 versus 3 
flagella. Histograms include 
representative colorized tem images of 
each h. pylori strain analyzed. 
 
Image provided by Ms. Laura Martinez. 
 
 
Roughly half of the world’s population is infected with the bacterium Helicobacter pylori (H. pylori) yet 
the vast majority of infections remain asymptomatic. However, chronic H. pylori infection significantly 
increases the risk for number of diseases, including gastric ulcers and gastric cancers. H. pylori is 
shaped like a corkscrew (helical shape) and it has been proposed that this helical shape is important 
for the bacteria to burrow through the protective mucus layer of the stomach. Consistent with this 
notion, previous work from the Fred Hutch Laboratory of Dr. Nina Salama (Human Biology, Public 
Health Sciences and Basic Sciences Divisions) not only identified genes required for H. pylori’s 
helical shape (csd genes) but also showed that csd mutants were less efficient in colonizing the 
stomach in a mouse model. However, to directly test the role of cell shape, one would have to 
compare motile behaviors of helical rod with straight rod H.pylori strains, as well as determine the 
extent of shape variation in wild-type isolates of H. pylori. A new study from the Salama Lab, led by 
graduate student Laura Martinez and published in Molecular Microbiology, combined live-cell 
imaging with mathematical modeling to demonstrate that cell shape and flagellum number contribute 
independently to promote H. pylori motility in viscous environments. 
December 20, 2015 SCIENCE SPOTLIGHT 
 
2 Volume 5, Issue 12 | Fred Hutchinson Cancer Research Center 
 
The authors first chose three wild-type H. pylori strains from symptomatic patients and then used 
imaging software to determine that all three displayed marked variation in all morphological 
parameters examined, including: cell length, side curvature, flagellum number, and flagellum length. 
Next, the investigators imaged hundreds of live individual bacteria in environments designed to 
mimic the gel-like consistency of the human gastric mucus. This analysis revealed that all three 
strains not only exhibited overlapping distributions of swimming speeds but also that the trajectory of 
individual bacteria was complex, wherein forward swimming bouts were punctuated by reorientations 
and reversals. To assess which aspects of motility, if any, were dependent on helical shape, the 
authors compared motile behaviors of a wild-type H. pylori strain with its straight rod mutant 
counterpart (Δcsd6) generated in the same genetic background. Their results showed that the Δcsd6 
mutant displayed a significant reduction in both the maximum speed and in the fraction of motile 
bacteria. Because flagella-dependent motility is important for infection in other species, the 
researchers queried the role of flagellum number on motility. To this end, they measured several 
motility parameters in H.pylori strains that had either more, less, or normal number of flagella. This 
analysis revealed that swimming speed, but not other aspects of motility, correlated with flagellum 
number. Finally, a mathematical modeling algorithm known as resistive force theory (RFT) was used 
to compare the relative contributions of helical cell shape and flagellum number on swimming speed. 
Their RFT model first assumed constant motor torque and then varied the motor torque dependent 
on flagellum number. The RFT analysis showed that a combination of cell shape and flagellum 
number were required to explain the difference observed experimentally. Said Ms. Martinez: "The 
helical shape of H. pylori is a defining feature of this bacterium. Efficient motility is essential for H. 
pylori to persistently colonize the gastric mucus layer and to reach its niche adhered to stomach 
epithelial cells and in the proximal mucus overlying these cells. Our lab has previously shown that H. 
pylori mutants with altered shape do not colonize well. In this study, we experimentally confirmed the 
long held hypothesis that helical shape promotes faster swimming speeds in viscous solutions like 
gastric mucus, where H. pylori lives. Surprisingly, the differences in speed we measured are pretty 
subtle. We also showed that you could get a similar speed boost by inducing expression of 4 instead 
of 3 flagella. The fact that changing flagellum number could compensate for loss of helical shape 
makes us wonder if there are other selective pressures beyond swimming speed impacting 
morphology. Current work is focusing on localizing wild-type and mutant bacteria within infected 
tissue to further explore how cell shape impacts colonization". In summary, this study demonstrated 
that cell shape and flagellum number make important contributions to H. pylori motility, but there are 
likely to be other selective pressures that sculpt these morphological parameters in vivo. 
Funding for this work was provided by the National Institutes of Health and the National Science 
Foundation. 
December 20, 2015 SCIENCE SPOTLIGHT 
 
3 Volume 5, Issue 12 | Fred Hutchinson Cancer Research Center 
 
Martinez LE, Hardcastle JM, Wang J, Pincus Z, Tsang J, Hoover TR, Bansil R, Salama NR. 2016. 
Helicobacter pylori strains vary cell shape and flagellum number to maintain robust motility in 
viscous environments. Mol Microbiol. 99(1):88-110. doi: 10.1111/mmi.13218. 
 
 


The twists and turns in the life of Helicobacter pylori

March 16, 2015 SCIENCE SPOTLIGHT 
 
1 Volume 5, Issue 3 | Fred Hutchinson Cancer Research Center 
 
More Evidence in Support of Oral Prep Safety 
March 16, 2015 
     L Pattacini 
The use of antiretroviral therapy as prevention in highly HIV-exposed populations has proven to be 
effective in protecting against HIV infection in several studies. The Partners PrEP study 
demonstrated the safety and efficacy of oral tenofovir disoproxyl fumarate (Viread®) alone or in 
combination with emtricitabine (Truvada®), both inhibitors of HIV reverse transcriptase, as compared 
to placebo. These drugs were tested in almost 5,000 serodiscordant couples from Kenya and 
Uganda. While a high protection rate was shown, a concern has been raised that the use of 
treatment as prevention could induce selection for resistant strains of HIV. 
To address this problem, Dr. Dara Lehman from the Human Biology Division at Fred Hutch applied a 
very sensitive method to detect resistance-linked mutations in all subjects who seroconverted during 
the study or were acutely infected at the time of study randomization. A total of 122 HIV 
seroconversions occurred during the study, of which 18 were determined to be acute seronegative 
HIV infections at study enrollment. The median time between the estimated date of HIV infection and 
drug discontinuation, when the presence of the drug together with the virus could select for resistant 
strains, was 45 days. Mutations associated with resistance to the PrEP drugs were detectable in 
19% of seroconverted subjects, but in the majority of these cases resistance mutations were present 
only at a very low frequency, less than 1% of an individual's viral population. Since resistance at 
<1% has not previously been associated with increased risk of treatment failure, this frequency was 
used as a cutoff for the study. Five seroconverters had one or more mutations associated with their 
PrEP regimen at a level higher than 1%: three were individuals with unrecognized acute infection at 
the time of PrEP randomization, and only two were subjects that became infected during the trial and 
had evidence of PrEP use during HIV infection. 
To answer the question of whether PrEP might induce resistance in those who become infected 
despite its use is imperative at this time, as the FDA recently approved PrEP use outside of clinical 
trials, where the time between HIV tests is likely to be longer than the monthly testing that occurred 
during the clinical trials. Thus, in those that become infected, the coinciding presence of the drug 
and the virus together is likely to be longer. The data reveal that resistance to PrEP is an unlikely 
event in subjects receiving it as prophylaxis, but can occur both in unrecognized acute infections as 
well as in breakthrough infections, and more frequently when the combination of tenofovir and 
emtricitabine is used. The frequency of mutations leading to resistance is lower than previously 
March 16, 2015 SCIENCE SPOTLIGHT 
 
2 Volume 5, Issue 3 | Fred Hutchinson Cancer Research Center 
 
calculated with mathematical models, and, together with the data proving a low frequency of 
seroconversion in PrEP users, this study supports the safety of PrEP use in situations of high 
exposure. The conclusions of the study are summarized in Dr. Lehman's words: " The small risk of 
resistance must be balanced with the high number of infections averted by PrEP use, a prevention 
strategy safe and efficacious for situations of high exposure to HIV."   
Lehman DA, Baeten JM, McCoy CO, Weis JF, Peterson D, Mbara G, Donnell D, Thomas KK, 
Hendrix CW, Marzinke MA, Frenkel L, Ndase P, Mugo NR, Celum C, Overbaugh J, Matsen FA, the 
Partners PrEP Study. 2015. Risk of Drug Resistance Among Persons Acquiring HIV Within a 
Randomized Clinical Trial of Single- or Dual-Agent Preexposure Prophylaxis. J Infect Dis. Epub 
ahead of print. 
 
 
Figure provided Dr. Dara Lehman. 
Frequency resistance-linked mutations in seroconverters with PrEP drugs detected 
during or after HIV acquisition. Higher levels are shown in darker blue colors. 
FTC/TDF, emtricitabine plus tenofovir disoproxil fumarate; TDF, tenofovir 
disoproxil fumarate alone. 
 


More Evidence in Support of Oral Prep Safety

Zentner, Gabriel

March 16, 2015 SCIENCE SPOTLIGHT 
 
1 Volume 5, Issue 3 | Fred Hutchinson Cancer Research Center 
 
Myc Requires Mondoa to Reprogram Cancer 
Metabolism 
March 16, 2015 
     GE Zentner 
Overexpression of Myc-family transcription factors (TFs) is a hallmark of the majority of human 
cancers (Dang, 2012).  Overexpression of Myc-family TFs leads to alterations in gene transcription 
that influence numerous cellular processes, including metabolism, to facilitate increased cellular 
growth.  Myc TFs function in the context of a larger transcriptional network in which they dimerize 
with the Max protein to bind specific DNA sequences, called E-boxes, to exert effects on gene 
expression.  Mxd proteins antagonize Myc transcriptional activity by competing for Max and E-box 
binding, and their expression is often correlated with arrested growth and cellular differentiation.  An 
additional Myc-like network, centered on the Mlx protein, operates in parallel  to the Myc network and 
involves dimerization with MondoA.  MondoA senses intermediates in glycolysis, which in turn 
promotes translocation of MondoA to the nucleus, where it increases the transcription of genes 
involved in glucose metabolism.  Interestingly, Mlx can associate with Mxd proteins, indicating a 
point of contact between the Myc arm of this network, which utilizes nutrients to promote cellular 
growth and proliferation, and the nutrient sensing abilities of MondoA.  To understand potential roles 
of Myc network TFs in facilitating cellular transformation by Myc proteins, postdoctoral fellows Dr. 
Patrick Carroll and Dr. Daniel Diolaiti in the lab of Dr. Robert Eisenman (Basic Sciences Division) 
and colleagues at the Universities of Washington and Utah undertook an siRNA screen of Myc 
network factors.  They found that depletion of MondoA, previously implicated in metabolic sensing, 
severely reduced the viability of Myc-overexpressing cells, suggesting a potential metabolic 
vulnerability of Myc-transformed cells. 
In an siRNA screen targeting Myc network members, the authors found that knockdown of MondoA, 
and to a lesser extent its dimerization partner Mlx, in the context of increased c-Myc levels strongly 
reduced the viability of mouse fibroblasts.  MondoA depletion also reduced the viability of several 
human cancer cell lines overexpressing c-Myc, indicating a general requirement for MondoA in c-
Myc-induced cell growth.  Extending these observations to n-Myc, which is overexpressed in ~25% 
of neuroblastomas, the authors found that MondoA knockdown strongly reduced viability of 
neuroblastoma cell lines with N-Myc overexpression but not those with normal N-Myc levels. 
To gain insight into the reliance of N-Myc overexpression on MondoA, the authors made use of a 
neuroblastoma cell line enabling inducible expression of N-Myc known as Tet21N.  Knockdown of 
March 16, 2015 SCIENCE SPOTLIGHT 
 
2 Volume 5, Issue 3 | Fred Hutchinson Cancer Research Center 
 
MondoA in Tet21N cells following N-Myc induction impaired proliferation and also increased 
apoptosis.  Furthermore, increased levels of the cell cycle inhibitors p21 and p53 were detected after 
MondoA knockdown, suggesting that MondoA dampens cellular stress associated with N-Myc 
overexpression to ensure continued cell survival and proliferation.  Indeed, MondoA knockdown 
impaired the ability of N-Myc-overexpressing Tet21N cells to form colonies in soft agar and to form 
tumors in mouse xenograft models. 
As N-Myc and MondoA are TFs, the authors next asked if loss of MondoA altered the transcriptional 
properties of N-Myc.  While MondoA knockdown did not substantially alter the set of genes altered 
by N-Myc, the magnitude of these changes was significantly attenuated.  N-Myc required MondoA 
for maximal expression of genes involved in promoting cell growth and proliferation, including many 
genes involved in nucleotide biosynthesis, amino acid metabolism, mitochondrial function, and lipid 
biosynthesis.  The expression of these genes was also dependent on the MondoA dimerization 
partner Mlx, indicating a requirement for MondoA-Mlx heterodimers in upregulating metabolism in 
the context of N-Myc overexpression.  Lastly, the researchers found that knockdown of a number of 
metabolic genes induced by MondoA substantially reduced cell viability in N-Myc overexpessing 
cells.  Metabolic profiling of neurolastoma cells revealed broad alterations in cellular metabolism 
upon N-Myc activation and shows that these cells depended on lipid biosynthesis for survival.  
To assess the relevance of metabolic upregulation to human cancers, the authors identified a gene 
expression signature consisting of Myc/MondoA-regulated metabolic genes and assessed the 
expression of this signature in patients with six distinct cancers: neuroblastoma, lung squamous cell 
carcinoma/lung adenocarcinoma, liver hepatocellular carcinoma, colon adenocarcinoma, acute 
myeloid leukemia, and breast invasive carcinoma.  High expression of the signature was significantly 
correlated with poor prognosis in all cases to differing degrees. 
This study defines a mechanism through which deregulated Myc is able to hijack cellular metabolism 
to promote tumorigenic cell growth and proliferation.  "Our synthetic lethal condition represents an 
imbalance of metabolic economics; a failure of supply in the context of insatiable (or increased) 
demand," said Dr. Carroll.  These findings also open new possibilities for therapy of Myc-
overexpressing cancers.  "Our findings have clear implications for the therapeutic treatment of Myc 
over-expressing malignancies and suggest that inhibiting MondoA/Mlx transcriptional activity or 
blocking lipid metabolism could be an effective therapeutic strategy to selectively target these 
cancers," said Dr. Diolaiti. 
 
 
March 16, 2015 SCIENCE SPOTLIGHT 
 
3 Volume 5, Issue 3 | Fred Hutchinson Cancer Research Center 
 
 
Carroll PA, Diolaiti D, McFerrin L, Gu H, Djukovic D, Du J, Cheng PF, Anderson S, Ulrich M, Hurley 
JB, Raftery D, Ayer DE, Eisenman RN. 2015. Deregulated Myc Requires MondoA/Mlx for Metabolic 
Reprogramming and Tumorigenesis. Cancer Cell27(2):271-285. 
Also see: Dang CV. 2012. MYC on the path to cancer. Cell 149(1):22-35. 
 
 
Image provided by Dr. Daniel Diolaiti 
Myc/Max and MondoA/Mlx, both separately and cooperatively, 
influence the expression of genes involved in the metabolism of the 
cell’s building blocks, leading to increased cell growth and 
proliferation. When MondoA is lost, fatty acid metabolism is 
impaired, leading to a failure of growth and subsequent cell death. 
 


Not a member yet