spotlight

Published online June 16, 2006 • 110.1021/cb600237e CCC: $33.50

© 2006 by American Chemical Society

264 ACS CHEMICAL BIOLOGY • VOL.1 NO.5 www.acschemicalbiology.org

Elusive Couple Finally Captured In the eukaryotic nucleus, RNA polymerase II (Pol II) and a vast collection of accessory proteins recognize DNA

elements and then initiate and transcribe pre-messenger RNAs (pre-mRNAs). Before an RNA is ripe for translation, it must be processed in a number of ways, including 5' capping, splicing, 3' cleavage, and polyadenylation. A wealth of recent studies indicates that these events are probably not as linear as once envisioned. Rather, a web of partnerships exists between the RNA synthesis machinery and various processing machines. Studying these connections has proved difficult for the biochemist because of the

lack of a reliable system where multiple activities can be observed. Now, two groups have optimized an efficient and reliable coupled-transcription/splicing system in a nuclear extract derived from a human cell line. Both groups show that the addition of a DNA template containing the proper sequence elements results in transcription as well as splicing.

Studies of splicing in a test tube have typically involved the addition of a radiolabeled RNA synthesized beforehand by a bacteriophage RNA polymerase. Hicks et al. (PLoS Biol. 2006, 4, e147) assemble reactions with template DNA containing either a promoter specific for the bacteriophage polymerase, T7, or the eukaryotic polymerase, Pol II. Adding exogenous T7 RNA polymerase to the nuclear extract made both promoters competent for transcription, but a striking difference was observed in splicing. Transcripts synthesized by Pol II entered the splicing pathway efficiently, whereas T7 transcripts accumulated but displayed little splicing. The Pol II transcripts also displayed greater resistance to nucleases. This resistance was dependent upon splicing signals in the pre-mRNA and on the integrity of U2 small nuclear RNA, a spliceosomal RNA employed early in the splicing pathway.

SpotlightChaos HelpsHybridization of DNA is a routine step in a wide variety of methods used to analyze gene sequences and expression levels. The development of DNA microarrays, where thou-sands of nucleic acid strands can be analyzed on the surface of a single glass microscope slide, has enabled rapid, large-scale detection of DNA. However, the hybridization step in the process is performed in a diffusion-limited manner, and the distance that DNA can diffuse in the time allotted for hybridization is an order of magnitude shorter than the typical length of most microarrays. Now, Liu et al. (Angew. Chem., Int. Ed. 2006, 45, 3618–3623) present a method to accelerate the DNA hybridization process by using microflu-idic chaotic mixing.

A microfluidic device was designed and constructed to facilitate effective circulation and mixing of the hybridization solution while maintaining compatibility with the microarray format. The silicone rubber device, which can be sealed onto a microarray slide, contains two symmetric hybridization chambers integrated with peristaltic pumps that circulate the fluid between the chambers. The chambers are connected by bridge channels, where herringbone patterns on the ceiling continuously introduce the chaotic mixing of the solution. Bifurcating channels equalize the mixed solution distributed into the chambers.

Assessment of the efficacy of the device demonstrated that hybridizations subjected to chaotic mixing resulted in

microarrays with stronger sig-nals, enhanced sensitivity,

reduced spot-to-spot variability, improved signal specificity, and an increase in molar-hybridization

events over static hybridizations. Notably,

hybridizations in which fluid was circulated but not subjected to chaotic mixing revealed that chaotic mixing was a significant contributor to the signal enhancement. The superior data quality, higher efficiency, and compatibility with current microarray protocols provided by the device champion its incorporation into microarray protocols. EG

Courtesy of Getty images

(continued on page 265)

Reprinted with permission from Angewandte Chemie, International Edition

Spotlight

265www.acschemicalbiology.org VOL.1 NO.5 • ACS CHEMICAL BIOLOGY

Tackling Tuberculosis

Finally, the researchers tested an alternative splicing pre-mRNA containing one set of 5' splicing signals, but two sets of 3' splicing signals. Interestingly, splice site choice was rather different when Pol II syn-thesis was feeding the pre-mRNA to the spliceosome.

In a similar effort, Das et al. (Genes Dev. 2006, 20, 1100–1109) optimize a coupled reaction and compare Pol II with T7-synthesized transcripts in their splicing fates. Again, the efficiency of splicing was far better with the eukaryotic polymerase driving pre-mRNA production. This study looked closely at spliceosomal complex formation on the pre-mRNA and at how the polymerase identity affected these events. Remarkably, the pre-mRNAs produced by Pol II were nearly quantitatively shuttled into spliceosomes as they were synthesized, whereas T7 transcripts remained in heterogeneous RNA–protein

complexes, which hinder splicing. This transfer from Pol II to the spliceosome depended upon intact splice site signals. Additionally, this study indicated that splic-ing kinetics are far more rapid with Pol II pre-mRNAs.

Both of these studies demonstrate biochemically that a functional handshake occurs between two nuclear machines, the Pol II complex and the spli-ceosome. With these reliable in vitro systems, the hands that do the shaking can finally be sought and characterized. Of particular interest will be the phos-phorylated carboxy-terminus of Pol II, a protein domain that seems to mediate crosstalk with the processing machines. The stark difference between Pol II and T7 polymerase lends itself nicely to various domain-swapping experiments. Finally, this powerful meth-odology will offer scientists new perspectives on the mysterious realm of alternative splice site choice. JU

Elusive Couple Finally Captured, continued

More than one-third of the world’s popu-

lation is infected with Mycobacterium

tuberculosis, the bacteria that causes

tuberculosis (TB), and the emergence of

strains that are resistant to antibiotic treat-

ment has fueled interest in finding new

drugs to combat the disease. The cell wall

of M. tuberculosis is a particularly attrac-

tive drug target because the organism

depends on the integrity of its cell wall for

survival. Moreover, many of the enzymes

used in cell-wall biosynthesis are not found

in humans, so potential drug side effects

are minimized. Rose et al. ( JACS 2006, 128,

6721–6729) report the expression, purifi-

cation, and characterization of a galactosyl-

transferase (GT) involved in M. tuberculosis

cell-wall biosynthesis.

The major structural component of the

M. tuberculosis cell wall is a lipidated

polysaccharide, the mycoyl–arabinogalactan–

peptidoglycan (mAGP) complex. The

carbohydrate-based core of mAGP is

composed of ~30 D-galactofuranose

residues attached via alternating β-(1 → 5)

and β-(1 → 6) linkages. A GT responsible

for constructing these linkages, termed

glfT, was expressed in Escherichia coli,

and several potential di- and trisaccharide

glycosyl acceptors were used to examine

the substrate specificity of the enzyme.

Characterization of the products by NMR

and MS revealed that the recombinant

enzyme, like the wild-type enzyme, is

capable of creating both β-(1 → 5) and

β-(1 → 6) linkages. In addition, both

disaccharides and trisaccharides were

substrates for the enzyme, but

kinetic analysis revealed that

the trisaccharides’ acceptors

were better substrates than those of the

disaccharides.

The insight into the biosynthesis of

the mAGP complex provided by this

study enabled the authors to propose

that ≤3 additional enzymes are required

to generate the substrate for glfT in vivo.

Further characterization of glfT and other

enzymes in the biosynthetic pathway will

facilitate discovery of inhibitors of M. tuber-

culosis cell-wall biosynthesis, paving the

way for new treatments for TB. EG

OO

HO

OH

O Galactan

OHO

HO

OHO

OHO

OHHO

HO

HOReprinted with permission from the Journal of the American Chemical Society

Spotlight


Microengraving and Multiplexing MonoclonalsMonoclonal antibodies have found powerful applica-

tions in therapeutics and as tools for detection of bio-

molecules. The process of identification and retrieval of

cells that produce antigen-specific antibodies, however,

is notoriously labor-intensive and time-consuming.

Now, Love et al. (Nat. Biotechnol. 2006, 24, 703–707)

describe a microengraving method that enables rapid

identification, recovery, and clonal expansion of mono-

clonal antibody-generating cell lines.

The microengraving method is designed so that

the secreted antibodies of individual cells arrayed in

high-density microwells are deposited on a microar-

ray surface. Subsequent analysis of the microarray

pinpoints the location of the corresponding antibody-

secreting cell, which can then be extracted from the

microwell and expanded in culture. The efficacy of the

microengraving approach was demonstrated when

cells were identified that produce antibodies to major

histocompatibility complex type I (H-2Kb).

Hybridoma cells were generated from

mice immunized with peptide-loaded

H-2Kb/streptavidin tetramers and arrayed

into microwells fabricated on slabs of

poly(dimethylsiloxane). The slabs were

sealed onto a microarray surface coated

with appropriate secondary antibody,

and secreted antibodies were captured

and detected with fluorescently labeled

H-2Kb. Out of 200,000 cells screened,

4300 positive spots were generated on the

microarray, and of the 50 corresponding cells that were

arbitrarily selected for expansion, 17 active hybridoma

supernatants were identified.

This approach dramatically improves several aspects

of monoclonal antibody generation. First, early screen-

ing of the cells eliminates the tedious serial dilution

process for achieving monoclonality and the need to

maintain several potential cell lines during antibody

Cell polarity is a complex pro-cess that requires the intricate coordination of many different biomolecular interactions. The Rho-GTPase Cdc42 is involved in maintaining polarity in epithelial cells, but the variety of processes that require Cdc42 has precluded a more precise understanding of its role in cell polarity. It is known that Cdc42 function is regulated by numer-ous guanine exchange factors (GEFs) and GTPase-activating proteins (GAPs), and now Wells et al. (Cell 2006, 125, 535–548) describe a protein interaction network surrounding the Cdc42 GAP Rich1 that links Cdc42 to cell polarity.

In order to find proteins that interact with Cdc42 in the con-text of cell polarity, the authors

characterization. In addition, the method facilitates isolation of

slow-growing and rare clones that are not easily accessed via

traditional methods. Finally, clones with unique specificities can

be rapidly identified when they are multiplexed with differen-

tially labeled antigens in the microarray analysis. This innovative

method could also have widespread applications in the screening

and characterization of cells that secrete other types of molecules,

such as cytokines. EG

PATJ

Rich1

AlphaAdducin

CIN85

Angiomotin

Angiomotin L1

Angiomotin L2

Par3 CD2AP

KIF3

Par6

LIN7

Crumbs

MUPP1

MPP7

CAPZα

CAPZβ

aPKC

PP2A-α1

MFAP1

Pals1

Repr

inte

d w

ith p

erm

issi

on fr

om C

ell

used FRET and tandem MS to screen 50 GEFs and 50 GAPs for their ability to alter the GTPase cycle of Cdc42 and to bind to known polarity proteins. The GAP Rich1 was found to associ-ate with polarity proteins at tight junctions (TJs) (complexes that form a seal between epithelial cells), and it was determined that the GAP activity of Rich1 is required for proper TJ mainte-nance, perhaps by preventing accumulation of Cdc42-GTP at the TJs.

Tandem MS was used to further probe the polarity protein interaction network to reveal additional Rich1 interactions. Rich1 associates with the scaffolding protein angiomotin (Amot) through the coiled-coil domain of Amot. In addition, Amot associates with Patj, a protein involved in establishing apical junctions, through the PDZ domain binding motif of Amot. This interaction targets Amot to TJs and links Rich1 to Patj, thereby targeting Rich1 to Cdc42 at TJs. The authors pro-pose that Rich1 and Amot help maintain TJ integrity by regulat-ing Cdc42 activity at TJs and by integrating Cdc42 function with the trafficking of intracellular polarity proteins at the TJ. EG

Polarity Networks Get Rich

Spotlight


Maintaining an organism’s energy balance and food intake is a critical func-tion of the central nervous system, the circuitry of which has been shown previ-ously to sense glucose and fatty acid levels directly. Emerging data also suggest that specific populations of neurons sense fuel availabil-ity through hormones such as leptin, a small protein known to influence hunger, food con-sumption, and energy expenditure. Now Cota et al. (Science 2006, 312, 927–930) show that the hypothalamus senses L-leucine and responds to changes in leptin levels. These responses are

Nonribosomal peptides (NRPs) are chemically diverse linear or cyclic molecules used extensively in therapeutics. NRPs are synthesized by unicellular organisms, plants, and fungi using a large enzyme complex with a common core structure and many different modules to chemically change the evolv-ing product. This synthetic mechanism has forced research-ers to rely on complex solid-phase synthetic schemes

to incorporate the variety of L-, D-, β-, N-methyl, and α,

β-unsaturated amino acids into NRPs. Seebeck and Szostak ( JACS 2006, 128, 7150–7151) now use a simple and versa-tile E. coli translation system to synthesize dehydroalanine-containing linear and cyclic peptides from messenger RNA templates.

Ribosomal peptide synthesis needs an appropriately charged transfer RNA (tRNA), which requires tRNA synthe-tases to use non-natural amino acids as substrates. The authors show that selenalysine is a substrate for lysine aminoacyl–tRNA–synthetase and can be incorporated into

peptides by a reconstituted translation system from E. coli. Oxidative elimination converts the linear peptide into dehydro-peptide, generating a scaffold with which many different nucleo-philes can react. The authors show that such model dehydro-peptides can also be synthesized as a stable cyclic structure. This simple ribosome-based system to make a reactive scaffold provides a mechanism to generate highly decorated peptides without the use of solid-phase synthesis. Furthermore, this synthetic scheme is amenable to high-throughput screening techniques and opens the door to the selection of bioactive peptide compounds. EJ

governed by the serine–threonine kinase mammalian target of rapamycin (mTOR).

In the presence of mitogens and nutrients, mTOR stimulates cell growth,

and aberrant over-activity of this protein has been associated with cancer, diabetes, and obesity. The authors observed that fasted rats exhibited decreased levels of phos-phorylated S6 kinase, a

target of mTOR. They then showed that treatment of fasted rats with L-leucine decreased the amount of food these rats ate and contributed to weight reduction. Furthermore, co-treatment of rats with L-leucine and the mTOR inhibi-tor rapamycin eliminated the L-leucine-

induced anorexia; this suggests that mTOR responds directly to amino acid levels.

The researchers further showed mTOR to be a nutrient sensor by examining the effect of the cyto-kine leptin on the hypothalamus. Treatment with leptin increased the levels of phosphorylated S6 kinase and induced anorexia; this suggests that mTOR is involved in this process. mTOR’s participation was confirmed by co-administra-tion of leptin and rapamycin, which reversed the anorexia.

Further studies of this and other fuel-sensing pathways may provide insight into the relation-ships between obesity and type 2 diabetes. ST

Signaling Brain-Energy Balance

Amino acids(L-leucine)

Leptin Re-feeding

Food intake

Rapamycin

mTOR AMPK

NH O

HN

NH

Se

O

HN

NH3

KSe Selenopeptide Dehydropeptide, ∆Ala

H3N

Se

NH3

O

O–

Reprinted with permission from the Journal of the American Chemical Society

Nonribosomal Peptides from the Ribosome

Repr

inte

d w

ith p

erm

issi

on fr

om A

AAS

Spotlight


Metabolizing MetabolitesThiamin pyrophosphate is an essential coenzyme

involved in the metabolism of carbohydrates and amino acids. The chemical structure of thiamin

consists of a thiazole portion and a pyrimi-dine portion. Although it is known that

thiamin biosynthesis in bacteria employs five different enzymes,

only thiazole synthase (Thi4) has been found in eukary-otes thus far. Chatterjee

et al. (JACS 2006, 128, 7158–7159) identify and char-

acterize a metabolite of Thi4 from Saccharomyces cerevisiae, providing

clues to the biosynthetic pathway of this important molecule in eukaryotes.

The role of Thi4 in thiamin biosynthesis is not known, but it was observed that denaturation of Thi4

resulted in the release of four major enzyme metabo-

Yes to NO DetectionNitric oxide (NO) is an important signal-ing molecule involved in many biological processes, including immune response, neurotransmission, and blood-pressure regulation. NO is one of the few known gaseous signaling molecules, and its direct detection in vivo has been notori-ously difficult because of its rapid dif-fusion and reactivity in the cell. Current NO detection methods are available, but they suffer from indirect measurement of NO or low spatial resolution, or they require complex instrumentation. Now, Lim et al. (Nat. Chem. Biol., published online May 28, 2006, doi: 10.1038/nchembio794) describe the synthesis, characterization, and biological applica-tion of a fluorescent probe that enables rapid, direct, and specific detection of NO in live cells.

lites. One of the metabolites was stable enough to be char-acterized, and several analytical tools were used to piece together its structure. Sequence analysis of the enzyme and UV absorption properties of the metabolite suggested that the compound was adenylated. HPLC, NMR, and MS experiments enabled the researchers to propose a struc-ture that contained an adenosyl and a thiazole moiety. To confirm the presence of the thiazole, the authors subjected the metabolite to a series of reactions that yielded an intensely fluorescent thiochrome phosphate, a compound that would be generated only if a thiazole was originally present. The elucidation of the metabolite’s structure allowed the authors to propose a mechanism for its biosynthesis wherein it is derived from NAD, and chem-istry similar to that involved in ADP ribosylation contributes to the creation of the thiazole functionality. These results demonstrate the feasibility of examining enzyme-bound metabolites to help decipher the function of enzymes with unknown activity. EG

The NO probe was designed so that exposure of a relatively nonfluo-rescent, cell-permeable reagent to NO results in the generation of a fluorescent compound that could be visualized in real time by fluorescence microscopy. To this end, a fluorescein derivative was reacted with CuCl2 to generate a Cu(II) fluorescein (CuFL) species with reduced fluorescence properties. Reaction of CuFL with NO results in a nitrosated fluores-cein (FL-NO) derivative. Characterization by EPR, UV–vis, NMR, fluorescence spec-troscopy, and LC/MS revealed that expo-sure of NO to CuFL induces the reduction of Cu(II) to Cu(I) and the concomitant

formation of NO+, which rapidly reacts to form FL-NO. Dissociation of FL-NO from the Cu(I) species results in the fluorescence increase.

The utility of CuFL to detect NO was tested in two cell lines. In human neuroblastoma cells, estrogen activates constitutive nitric oxide synthases (NOSs) to produce NO. Simultaneous addition of estrogen and CuFL to these cells resulted in an NO-dependent fluorescence response. Similarly, induc-ible NOSs are activated in macrophages treated with lipopolysaccharide and interferon-γ, and addition of CuFL to these cells also resulted in an NO-dependent increase in fluorescence. These results demonstrate the power of CuFL as a direct NO-detection reagent for a variety of applications. EG

Repr

inte

d w

ith p

erm

issi

on fr

om N

atur

e Ch

emic

al B

iolo

gy

Reprinted w

ith perm

ission fr

om the Jo

urnal o

f the Americ

an Chemical S

ociety

Spotlight


Bioorganic Chemistry, GRC July 30–August 4, 2006 Oxford, U.K.

Protein Phosphorylation and Cell SignalingAugust 18–22, 2006 La Jolla, CA

Annual Meeting of the Protein Society August 5–9, 2006 San Diego, CA

Multiprotein Complexes Involved in Cell Regulation, KeystoneAugust 18–23, 2006St. John’s College, U.K.

Post-Transcriptional Gene Regulation, GRC August 13–18, 2006 Oxford, U.K.

Translational Control, CSHLSeptember 6–10, 2006Cold Spring Harbor, NY

UPCOMING CONFERENCES

Spotlights written by Eva Gordon, Evelyn Jabri, Sarah Tegen, and Jason Underwood

Teaching Old Drugs New TricksBringing a new drug to market takes nearly

15 years and almost $1 billion. Efforts to

speed up the process and decrease the costs

have led researchers to search for new appli-

cations for known drugs. Now, Chong et al. ( J. Med. Chem.

2006, 49, 2677–2680) report the screening of a library

of 2450 known drugs for new angiogenesis inhibitors,

identify the immunosuppressant mycophenolic acid (MPA)

as an inhibitor of endothelial-cell growth, and determine

its mechanism of action and efficacy as an antiangiogenic

agent.

MPA is commonly used to prevent organ-transplant

rejection, and its activity as an inhibitor of inosine

monophosphate dehydrogenase (IMPDH), an enzyme

involved in de novo guanine biosynthesis, is well-estab-

lished. However, the mechanism of action and molecular

target of MPA as an angiogenesis inhibitor needed to be

determined. Examination of the effects of MPA on cell

proliferation and cell-cycle progression in human umbili-

cal vein endothelial cells (HUVECs) indicated that, as in

T and B cells, MPA inhibits HUVEC growth in a guano-

sine-dependent manner and causes cell-cycle arrest in

G1. Moreover, examination of the effects of selectively

knocking down both isoforms

of IMPDH (1 and 2) in HUVEC by

RNA interference revealed that

loss of IMPDH-1 function is suffi-

cient to cause cell-cycle arrest in

G1. In contrast, knocking down

IMPDH-1 has no effect on T-cell

proliferation. Further, an in vivo

angiogenesis assay demonstrated that mice treated with

MPA had significantly less new blood vessel formation

than control mice. Finally, a murine renal cell carcinoma

model was used to determine that MPA prevents tumor-

induced angiogenesis. Taken together, the data validate

IMPDH-1 as the MPA target in endothelial cells, and given

that IMPDH-2 appears to be the more essential protein

for T-cell proliferation and development, the drug-target

potential of IMPDH-1 is substantiated. These results high-

light the power of this drug-discovery strategy and point

to specific inhibitors of IMPDH-1 as possible antiangio-

genic drugs. The library of existing drugs reported in this

paper will be made available to the scientific community

for screening on other drug targets through collaborative

agreements or other mutually agreeable arrangements. EG

Repr

inte

d w

ith p

erm

issi

on fr

om th

e Jo

urna

l of M

edic

inal

Che

mis

try

spotlight

Documents