DOI: 10.1126/science.1223811, 1120 (2012);336 Science

Oliver M. Bannard and Jason G. CysterWhen Less Signaling Is More

This copy is for your personal, non-commercial use only.

clicking here.colleagues, clients, or customers by , you can order high-quality copies for yourIf you wish to distribute this article to others

  here.following the guidelines

can be obtained byPermission to republish or repurpose articles or portions of articles

  ): November 11, 2014 www.sciencemag.org (this information is current as of

The following resources related to this article are available online at

http://www.sciencemag.org/content/336/6085/1120.full.htmlversion of this article at:

including high-resolution figures, can be found in the onlineUpdated information and services,

http://www.sciencemag.org/content/336/6085/1120.full.html#relatedfound at:

can berelated to this article A list of selected additional articles on the Science Web sites

http://www.sciencemag.org/content/336/6085/1120.full.html#ref-list-1, 4 of which can be accessed free:cites 13 articlesThis article

http://www.sciencemag.org/cgi/collection/immunologyImmunology

subject collections:This article appears in the following

registered trademark of AAAS. is aScience2012 by the American Association for the Advancement of Science; all rights reserved. The title

CopyrightAmerican Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005. (print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by theScience

on

Nov

embe

r 11

, 201

4w

ww

.sci

ence

mag

.org

Dow

nloa

ded

from

o

n N

ovem

ber

11, 2

014

ww

w.s

cien

cem

ag.o

rgD

ownl

oade

d fr

om

on

Nov

embe

r 11

, 201

4w

ww

.sci

ence

mag

.org

Dow

nloa

ded

from

1 JUNE 2012 VOL 336 SCIENCE www.sciencemag.org 1120

PERSPECTIVES

When Less Signaling Is More

IMMUNOLOGY

Oliver M. Bannard and Jason G. Cyster

Negative regulation of B cell receptor signaling

may contribute to B cell selection and cell fate

determination in germinal centers.

Constrained signaling. Signaling by the B cell receptor is restricted by phosphatases (such as SHP-1) in G1 phase of GC B cells, with possible effects on B cell fate. MHC II, major histocompatibility complex II.

CR

ED

IT: B

. S

TR

AU

CH

/SC

IEN

CE

A crowded party means lots of sensory

inputs and the need to “tune out”

distracting signals to have produc-

tive interactions. On page 1178 in this issue,

Khalil et al. ( 1) suggest that in a crowded

environment, B cells too must tune out dis-

tracting inputs so that they can be responsive

to key interactions.

Antibody aff inity maturation occurs

inside germinal centers (GCs) within lym-

phoid tissues and involves B cells testing

newly mutated B cell receptors (BCRs) for

improved binding to the stimulatory anti-

gen. Only the rare mutant cells that bind bet-

ter than their sister cells survive, and can

either undergo further rounds of mutation

and selection, or they can differentiate into

antibody-secreting plasma cells or memory

cells. But how is antigen binding by the BCR

“read out” so that only the high-affi nity cells

survive? Two nonexclusive models have

been considered: High-affi nity cells transmit

“winning” BCR signals; and high-affi nity

cells internalize and present more antigen,

receiving “winning” T cell help ( 2, 3).

To test the fi rst model, BCR signaling by

GC B cells must be studied. Yet, in contrast

to non-GC B cells, investigation of signaling

in GC B cells has been limited because of

their rarity and propensity to die within hours

of isolation. The study of Khalil et al. begins

to address this important gap in knowledge.

An early step in BCR signaling in non-

GC B cells is activation of tyrosine kinases,

including Syk, and phosphorylation of adap-

tor molecules such as BLNK. GC B cells

repeatedly encounter cognate antigen, so they

might be expected to have elevated “base-

line” BCR activity. To examine this possi-

bility, Khalil et al. applied an “instant-fi x”

approach whereby mouse spleen cells were

isolated directly into fi xative to capture sig-

naling molecules in their in vivo phosphoryl-

ation state. Unexpectedly, the amounts of

phosphorylated Syk, BLNK, and total phos-

phorylated tyrosine residues were lower in

GC compared to non-GC B cells—even after

stimulation via surface immunoglobulin M

(IgM). The lower amount of surface BCR on

GC B cells might contribute to reduced sig-

naling, but comparison of cells with matched

amounts of IgM suggested additional con-

straints on the GC cells. Indeed, when pro-

tein tyrosine phosphatase (PTPase) function

was globally inhibited, the amounts of phos-

phorylated Syk and BLNK increased in GC

B cells. Surprisingly, in IgM-stimulated GC

B cells, the BCR and SHP-1, a major cyto-

solic PTPase, were codistributed, whereas in

non-GC B cells they were segregated (see the

fi gure). Temporal deletion of SHP-1 from B

cells led to a fi vefold reduction in GC size.

Whether this loss was due to exaggerated

BCR signaling or reflected a subsequent

defect—for example, in antigen presenta-

tion—is not yet clear.

In a fi nal twist, the authors noted that a

minor population of cells expressing high

levels of phosphorylated Syk was detect-

able in the germinal center. They specu-

lated that this might relate to cell division

cycle status and indeed found that cells in G2

phase showed a reduced amount of SHP-1

and reduced colocalization of SHP-1 and

the BCR. Thus, phosphorylation-dependent

BCR signaling may be reactivated tran-

siently during the G2 phase of each cell cycle.

In its simplest interpretation, suppression

of BCR signaling during the main (G1) phase

of the cell cycle seems to favor the T cell–

based selection model, where GC B cells

with higher-affi nity BCRs endocytose more

antigen and receive “winning” amounts of T

cell help. But doesn’t receptor internalization

depend on signaling? Possibly not, at least

not on tyrosine phosphorylation of the BCR

and associated molecules that may be most

strongly suppressed by SHP-1. Indeed, the

BCR complexes internalizing with antigen

could be the ones with the least phosphoryl-

ation ( 4), although Syk-dependent signaling

may be required for antigen presentation by

B cells ( 5). However, neither of these studies

involved GC B cells, which may well differ

from non-GC B cells in their antigen-presen-

tation requirements ( 6).

Although Khalil et al. provide evidence

for increased negative regulation of BCR

signaling in GC B cells, a contribution

of BCR signaling to affinity-based selec-

tion is not yet ruled out. For example, dur-

Howard Hughes Medical Institute and Department of Micro-biology and Immunology, University of California–San Fran-cisco, CA 94143–0414, USA. E-mail: jason.cyster@ucsf.edu

Published by AAAS

www.sciencemag.org SCIENCE VOL 336 1 JUNE 2012 1121

PERSPECTIVES

Evidence of Things Not Seen

ASTRONOMY

Norman W. Murray

The Kepler space telescope can determine

the mass and orbital period of unseen planets

orbiting distant stars.

ing encounter with antigen in the germinal

center, engagement of the CD19 co-recep-

tor complex may occur. CD19 defi ciency is

associated with a defect in the GC response

( 7, 8). Perhaps CD19-mediated amplifi ca-

tion of BCR signaling overcomes the phos-

phatases to infl uence GC selection events. As

for the elevated BCR signaling during G2 ( 1),

cells in this phase may be enriched in the GC

dark zone ( 3), a region low in foreign antigen

and helper T cells ( 2), which might facilitate

negative selection of cells that have acquired

affi nity for ubiquitous self-antigens ( 8).

Many GC B cells undergo isotype switch-

ing, changing the class of antibody produced.

Whether the fi ndings of Khalil et al. will hold

for isotype-switched B cells will require fur-

ther studies. IgG and IgE have longer cyto-

plasmic domains than IgM that include a sig-

naling motif ( 9). Switching to IgE favors GC

B cell differentiation to short-lived plasma

cells ( 10). The relieved BCR signaling dur-

ing G2 might also enable the cell to sense

what isotype it is expressing and adjust its

cell fate decision-making accordingly.

What might be the basis for the distinct

partitioning of SHP-1 in non-GC and GC B

cells? The transmembrane protein-binding

partners of SHP-1 likely contribute to this

process, perhaps redundantly ( 11). Visual-

ization of SHP-1 in non-GC B cells showed

that stimulation via IgM is associated with

increased SHP-1 and BCR colocalization

( 12); whether this difference was due to the

use of distinct BCR-crosslinking agents, spe-

cifi c properties of the transgenic B cells stud-

ied by Khalil et al., or other factors merits

further investigation. SHP-1 also has affi n-

ity for membrane lipids and cytoskeletal pro-

teins ( 13). Clearly, more work is needed to

elucidate how SHP-1 is differentially regu-

lated in non-GC and GC B cells.

Why might there be such a big difference

in the BCR signaling requirements of non-GC

and GC B cells? One possibility is that reduced

BCR signaling actually augments antigen pre-

sentation. Another is that naïve B cells may

depend on strongly amplifi ed BCR signals to

bring them out of quiescence and ready them

for cell cycle; GC B cells are already in cycle,

perhaps minimizing the amount or type of sig-

naling needed for the BCR to infl uence cell

fate. Although Khalil et al. provide important

new insight regarding signaling in GC B cells,

further questions will need to be addressed if

we are to better grasp how antibody affi nity

maturation takes place.

References

1. A. M. Khalil et al., Science 336, 1178 (2012); 10.1126/science.1213368.

2. C. D. Allen et al., Immunity 27, 190 (2007). 3. G. D. Victora et al., Cell 143, 592 (2010). 4. P. Hou et al., PLoS Biol. 4, e200 (2006). 5. D. Le Roux et al., Mol. Biol. Cell 18, 3451 (2007). 6. N. A. Draghi, L. K. Denzin, Proc. Natl. Acad. Sci. U.S.A.

107, 16607 (2010). 7. Y. Wang, R. H. Carter, Immunity 22, 749 (2005). 8. C. C. Goodnow et al., Nat. Immunol. 11, 681 (2010). 9. N. Engels et al., Nat. Immunol. 10, 1018 (2009). 10. Z. Yang, B. M. Sullivan, C. D. Allen, Immunity, 10.1016/

j.immuni.2012.02.009 (2012). 11. L. Nitschke, Immunol. Rev. 230, 128 (2009). 12. H. Phee et al., Mol. Cell. Biol. 21, 8615 (2001). 13. U. Lorenz, Immunol. Rev. 228, 342 (2009).

10.1126/science.1223811

The inventory of known extrasolar plan-

ets (planets orbiting stars other than

our Sun) has grown explosively in the

past 3 years. The explosion relies on detect-

ing the small but distinct decrease in the fl ux

of light from a star caused by the passage of a

planet across the stellar disk, an event known

as a transit (see the fi gure). NASA’s Kepler

telescope, with a photometric precision of

~20 parts per million, has identified more

than 2300 extrasolar planetary candidates

( 1). For comparison, radial velocity surveys,

which identify planets by the orbital veloc-

ity they impart to their host stars, have found

only about a thousand planets since 1995.

For the brightest host stars, one can obtain

the high-quality spectra that are needed for

radial velocity measurements to verify or

exclude the planetary nature of the transiting

system. Unfortunately, the bulk of the transit

host stars are too dim to obtain radial veloc-

ities. Enter transit timing variations (TTVs)

in the length of a candidate extrasolar plan-

et’s year. Since the advent of Kepler, it has

become common to use TTVs to verify the

planetary nature of transiting objects in mul-

tiplanet systems ( 2, 3). Recently, TTVs were

used to infer the presence of an unseen plan-

etary companion, but with only weak con-

straints on the orbital period and companion

mass ( 4). On page 1133 of this week’s issue,

Nesvorný et al. ( 5) take the next step by iden-

tifying and characterizing unseen (nontran-

siting) planets using TTVs.

Transits have long fascinated astrono-

mers. Jeremiah Horrocks predicted and then

observed the transit of Venus in 1639. Tran-

sits of Venus, as seen from Earth, happen

rarely. The next transit will occur on 5 June

2012 and then again in 2117. The same dis-

tant observer on a planet orbiting a distant

star, in contrast, might see a transit every

Venusian year (~225 days), if they lie very

close to the plane of Venus’ orbit.

Distant observers might or might not see

transits of Earth, depending again on how

close they are to Earth’s orbital plane. How-

ever, even if they could not see transits of

Earth, they could readily detect the gravita-

tional effects of Earth on Venus (and the Sun).

The length of the Venusian year would be

seen to vary by 10 min as the Earth alternately

pulls Venus forward in its orbit and then drags

it back ( 6).

The analysis of publicly available data

for Kepler-object-of-interest 872 (KOI-872)

by Nesvorný et al. differs dramatically from

that published by the Kepler team. The latter

had previously identifi ed a 33.6-day-period

transiting planet with a radius similar to that

of Jupiter and a mass less than six Jupiter

masses (<6 MJ). Nesvorný et al. argue that

the system harbors two additional planets, a

57-day-period object not seen to transit the

star—inferred from the TTVs of the 33.6-day

object, with an estimated mass of 0.375 MJ—

and a second, previously unreported, transit-

ing object with a period of 6.77 days and a

radius about twice that of Earth.

The novelty of Nesvorný et al.’ s work

resides in the identifi cation of the nontransit-

ing body’s mass and orbital period. The use

of TTVs to fi nd unseen planets, although pre-

dicted some 7 years ago ( 6, 7), has not yielded

secure detections before this work.

Are the TTVs due to a second planet or

some other cause? Nesvorný et al. argue

rather convincingly that the former is the case,

although their claims will likely face intense

scrutiny. Better yet, they make predictions for

Canadian Institute for Theoretical Astrophysics, University of Toronto, 60 St. George Street, Toronto, ON M5S 3H8, Canada. E-mail: murray@cita.utoronto.ca

Published by AAAS