Subunit 1 of the Prefoldin Chaperone Complex Is Required for Lymphocyte Development and Function

This information is current as Shang Cao, Gianluca Carlesso, Anna B. Osipovich, Joan of October 1, 2021. Llanes, Qing Lin, Kristen L. Hoek, Wasif N. Khan and H. Earl Ruley J Immunol 2008; 181:476-484; ; doi: 10.4049/jimmunol.181.1.476 http://www.jimmunol.org/content/181/1/476 Downloaded from

References This article cites 50 articles, 18 of which you can access for free at: http://www.jimmunol.org/content/181/1/476.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication by guest on October 1, 2021 *average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2008 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Subunit 1 of the Prefoldin Chaperone Complex Is Required for Lymphocyte Development and Function1

Shang Cao,2 Gianluca Carlesso,2,3 Anna B. Osipovich, Joan Llanes, Qing Lin, Kristen L. Hoek, Wasif N. Khan,4 and H. Earl Ruley4

Prefoldin is a hexameric chaperone that facilitates posttranslational folding of actins and other cytoskeletal proteins by the Tcp1-containing ring complex chaperonin, TriC. The present study characterized mice with a null mutation in Pfdn1, which encodes the ﬁrst subunit of the Prefoldin complex. Pfdn1-deﬁcient mice displayed phenotypes characteristic of defects in cytoskeletal function, including manifestations of ciliary dyskinesia, neuronal loss, and defects in B and T cell development and function. B and T cell maturation was markedly impaired at relatively early stages, namely at the transitions from pre-pro-B to pre-B cells ,in the bone marrow and from CD4؊CD8؊ double-negative to CD4؉CD8؉ double-positive T cells in the thymus. In addition mature B and T lymphocytes displayed cell activation defects upon Ag receptor cross-linking accompanied by impaired Ag Downloaded from receptor capping in B cells. These phenotypes illustrate the importance of cytoskeletal function in immune cell development and activation. The Journal of Immunology, 2008, 181: 476–484.

ctins and tubulins, as ubiquitous components of the cy- protein substrates directly, Prefoldin signiﬁcantly enhances the rate toskeleton, play essential roles in fundamental cellular of TRiC-mediated protein folding (13). processes including motility, macromolecular transport, Unlike TRiC, yeast Prefoldin (GimC) is not essential for cell

A http://www.jimmunol.org/ signal transduction, and cell division (1). Actin and tubulin sub- viability (9, 10). However, GimC mutations generate cytoskeletal units assemble in a head-to-tail manner by a regulated, energy- defects similar to temperature-conditional TRiC mutations (9, 10), consuming process (2, 3). Fiber asymmetry enables actin and tu- providing further evidence for biochemical cooperation between bulin cytoskeletons to orchestrate rapid and spatially localized the chaperone and chaperonin. Prefoldin subunits 2, 3, and 6 in- changes in cell structure and function involved in diverse pro- ﬂuence Caenorhabditis elegans development (14, 15); thus, de- cesses from axon outgrowth to immune synapse signaling (4, 5). spite evidence of limited promiscuity among chaperones with re- Efﬁcient synthesis of actin and tubulin subunits requires two gard to substrate binding and chaperonin presentation (16), protein complexes, Prefoldin and TRiC/CCT (Tcp1 ring complex/ nonredundant activities of Prefoldin appear to be required for cy-

chaperonin-containing Tcp1) that facilitate protein folding (re- toskeletal function in lower eukaryotes. by guest on October 1, 2021 viewed in Ref. 6). Prefoldin is a molecular chaperone consisting of Genetic analysis of mammalian Prefoldin is expected to gener- six radially organized proteins, each forming a coiled-coil tentacle ate phenotypes indicative of biologically important folding sub- that participates in substrate binding (7, 8). Prefoldin was first strates in mammalian cells. In particular, we would like to know identified from genetic interactions implicating the complex in cy- whether Prefoldin mutant phenotypes are consistent with aberrant toskeleton biosynthesis (9) and as a biochemical activity that trans- cytoskeletal function, as in yeast, or possess defects involving a ferred denatured ␤-actin to TRiC (10). TRiC, a cytoplasmic chap- wider range of protein substrates. Mice deficient in Prefoldin chaperonin, transiently interacts with ϳ10% of cellular proteins and erone functions may also clarify the involvement of the cytoskel- folds several essential proteins, including actins and tubulins, by eton (17, 18) or other misfolded proteins in aging-associated dis- an ATP-dependent mechanism (reviewed in Ref. 11). Nascent eases (19–22). Finally, the availability of Prefoldin-deficient cells ␤-actin binds Prefoldin during translation (12) and is presented to and animals is expected to facilitate functional studies of the chap- the TRiC chaperonin complex (8, 10). Although TRiC can bind erone in different biological processes. In the present study, we characterized a null mutation in the gene (Pfdn1) for Prefoldin subunit 1 induced by gene trapping in Department of Microbiology and Immunology, Vanderbilt University, School of 5 Medicine, Nashville, TN 37232 mouse embryonic stem (ES) cells. We show that Pfdn1-deficient Received for publication January 2, 2008. Accepted for publication April 30, 2008. mice develop to term but display severe abnormalities character- The costs of publication of this article were defrayed in part by the payment of page istic of defects in cytoskeletal function, including ciliary dyskine- charges. This article must therefore be hereby marked advertisement in accordance sia, loss of neuron tracts in the brain, and defects in B and T cell with 18 U.S.C. Section 1734 solely to indicate this fact. development and function similar to but more severe than Wiskott- 1 This work was supported by Public Health Service Grants P01HL68744 (to H.E.R.) Aldrich syndrome protein (WASP) and WASP-interacting protein and AI060729-01 (to W.N.K.). K.L.H. was supported by National Institutes of Health Grant F32-AI069770-01. Additional support was provided by the Department of Mi- (WIP) mutations. These phenotypes appear to reflect processes crobiology and Immunology of Vanderbilt University and by Cancer Center Support such as immune cell signaling and axon guidance that place par- Grant P30CA68485 to the Vanderbilt-Ingram Cancer Center. ticular demands on cytoskeletal synthesis and remodeling. 2 S.C. and G.C. share first authorship and contributed equally to this work. 3 Current address: MedImmune, Inc., One MedImmune Way, Department of Inflam- mation and Autoimmunity, Gaithersburg, MD 20878 5 Abbreviations used in this paper: ES, embryonic stem; BM, bone marrow; MEF, 4 Address correspondence and reprint requests to Dr. Wasif N. Khan and Dr. H. Earl mouse embryonic fibroblast; MZ, marginal zone; WASP, Wiskott-Aldrich syndrome Ruley, Department of Microbiology and Immunology, Vanderbilt University, School protein; WIP, WASP-interacting protein; MFI, mean fluorescent intensity. of Medicine, Medical Center North, 1161 21st Avenue South, Nashville, TN 37232. E-mail addresses: [email protected] and [email protected] Copyright © 2008 by The American Association of Immunologists, Inc. 0022-1767/08/$2.00 www.jimmunol.org The Journal of Immunology 477 Downloaded from

FIGURE 2. Growth and survival of Pfdn1 mutant mice. A, Appearance of wild-type (left) and mutant (right) mice at 3 wk of age. B, Average weights of wild-type and heterozygous mice (gray) and homozygous mu- http://www.jimmunol.org/ tant mice (black) at 2 to 4 wk of age. C, Fraction of wild-type and heterozygous mice (gray) and homozygous mutant mice (black) surviving through 5 wk of age.

upstream (U) and downstream (D) primers 5Ј-CAGTCCTCCGATT FIGURE 1. Pfdn1 gene disruption. A, The GTR1.3 gene trap retrovirus GACTGAG-3Ј (Dp) and 5Ј-GGGGTTGTGGGCTCTTTAT-3Ј (Up). The was inserted into the third intron of the Pfdn1 gene as shown. The insert amplification protocol consisted of 35 cycles incubating at 94°C for 1 min, by guest on October 1, 2021 enables upstream Pfdn1 sequences to splice to a 3Ј exon consisting of the 55°C for 1 min, and 72 °C for 1 min. Mice and cells were genotyped by PCR analysis. Twenty to 50 ng of 3Ј end of a puromycin resistance gene (3Ј Puro), a ␤-galactosidase reporter genomic DNA (g) was used as template in a 50-␮l reaction using the Roche (lacZ), and a polyadenylation signal (PA), whereas Neo sequences are PCR mix with the upstream (U) and downstream (D) primers 5Ј-TGG expressed from the promoter of the RNA polymerase 2 gene (Pol2) to GATAATGCCCACAGGTA-3Ј (Dg), 5Ј-AAGCACTCAGAGCAGCA splice to exon 4. Primers complementary to proviral (Dp and Up) and AGTT-3Ј (Ug), and 5Ј-CAGTCCTCCGATTGACTGAG-3Ј (Dp). Samples cellular (Ug and Dg) sequences used for genotyping are indicated (where were amplified through 30 cycles of incubation at 94°C for 1 min, 55°C for D is downstream, U is upstream, p is provirus DNA, and g is genomic 1 min, and 72°C for 1 min. DNA). B, Sequence of GTR1.3-Pfdn1 fusion transcripts. Cellular sequences appended to Neo fusion transcripts were cloned by 3Ј RACE and Analysis of Pfdn1 gene expression sequenced, revealing an insert in the Pfdn1 gene. C, Genomic sequences The expression of Pfdn1 mRNA transcripts in mice and cells was assessed upstream of the GTR1.3 provirus were amplified by inverse PCR and se- by Northern blot hybridization to a [32P]dCTP-labeled probe derived from quenced, providing the exact location of the provirus in the Pfdn1 gene. D, the full length Pfdn1 cDNA, as described previously (24). Genotype analysis. Wild-type (ϩ/ϩ), heterozygous (ϩ/Ϫ), and homozygous Prefoldin 1 protein expression in ES cell clones was assessed by West- mutant (Ϫ/Ϫ) mice were genotyped by PCR using a mixture of Ug, Dg, and ern blot analysis. Cell lysates were fractionated by electrophoresis on a Dp primers. M, Marker. E, Western blot analysis of Pfdn1 expression. Em- 12% SDS-polyacrylamide gel and transferred onto a polyvinylidene diflu- oride membrane (PerkinElmer Life Science). Prefoldin 1 proteins were bryonic fibroblasts from mice with the indicated genotypes were analyzed by detected by using donkey polyclonal Abs raised against an internal region Western blot analysis using Abs against Pfdn1 (left panel) or heterogenous of the protein (Santa Cruz Biotechnology), as described previously (24). nuclear ribonuclear protein (hnRNP) C (right panel). F, Northern blot analysis of Pfdn1 expression. Ten micrograms of RNA from mouse embryonic fibro- Flow cytometric analysis blasts with the indicated genotypes were probed with Pfdn1-specific (left Flow cytometric analyses were performed for the phenotypic characteriza- panel)orGapdh-specific (right panel) sequences. tion of B and T lymphocytes as described (25, 26). Single-cell suspensions were prepared from lymphoid organs, and T cell populations in the thymus and spleens were identified by anti-CD4, anti-CD8, and anti-TCR␤. B cell Materials and Methods subpopulations in the bone marrow (BM) and spleen were identified by Gene trapping staining with anti-BP-1 (Ly-51), anti-CD25, anti-HSA, biotinylated anti- CD43 (S7), anti-B220, anti-CD23, anti-CD21, anti-CD19, anti-IgM, anti- The construction of a GTR1.3 retrovirus gene trap library, as well as the IgD, anti-CD5 (Ly-1), and streptavidin-allophycocyanin (purchased from analysis of 3Ј RACE products, has been described (23). BD Pharmingen) and anti-AA4.1 (purchased from eBiosciences). Data The insertion site was located by inverse PCR. Briefly, 50–200 ng of were collected on an BD LSR II cytometer (BD Biosciences) and analyzed genomic DNA from mouse tail was digested by AvaII (New England Bio- using FlowJo (Tree Star). labs) and 20–50 ng from the digest was ligated in a 20-␮l reaction (T4 For in vitro activation assays, splenocytes were isolated by hypotonic DNA ligase and 10ϫ ligase buffer with 1 mM ATP (New England Bio- lysis of RBC and cultured at 1 ϫ 106 cells/ml in RPMI 1640 medium labs)) at room temperature for 10 min. Two microliters of the ligation supplemented with 10% FBS (HyClone), 2 mM glutamine, 50 ␮M 2-ME, ␮ Ј product (p) was amplified by PCR in 50- l reaction (Roche) with the and 100 U/ml penicillin/streptomycin with or without F(ab )2 goat 478 PREFOLDIN FUNCTION IN LYMPHOCYTES

FIGURE 3. Histology of Pfdn1-deﬁcient Downloaded from mice. A–D, CNS defects. H&E-stained coro- nal sections through the cerebrum of wild- type (A) and homozygous mutant (B) mice and sagittal sections through the cerebellum of wild-type (C) and mutant (D) mice. The corpus callosum (CC) and arbor vitae (av) are indicated. E and F, Sinus infections. The http://www.jimmunol.org/ sinuses of mutant (F) but not wild-type (E) mice were plugged with mucus and often showed signs of massive infection. G and H, Spleen abnormalities. Sections of spleens (8 ␮M) from wild-type (G) and mutant (H) mice stained with H&E. Spleens from mutant mice were smaller and displayed disorganized red pulp and follicles. by guest on October 1, 2021

anti-mouse IgM (Jackson ImmunoResearch) and anti-CD3 for 18 h. Cells tured in DMEM (Mediatech), supplemented with 10% FCS (heat inacti- were also incubated with PMA and ionomycin (100 nM each) as positive vated at 55°C for 30 min) and 100 U/ml penicillin-streptomycin (Invitro- controls. B cell activation was determined by cell surface staining with gen). Primary mouse embryonic fibroblasts (MEFs) were cultured to near- anti-CD19 (allophycocyanin) and anti-CD86 (FITC). T cell activation was confluence densities and passed at 1:2 ratios until cell lines were obtained. determined by cell surface staining with anti-CD4 (FITC) and anti-CD69 (PE). Stained cells were incubated with 7-aminoactinomycin D (Molecular B cell receptor capping analyses Probes) immediately before acquisition, enabling the resolution of live vs Ϫ Ϫ Heterozygous control and Pfdn1 / splenocytes (1 ϫ 106) were stimulated dead cells by flow cytometry. Data were collected on an BD LSR II cy- with 10 ␮g/ml Cy5-conjugated goat anti-mouse (F(abЈ) ) IgM (Jackson tometer (BD Biosciences) and analyzed using FlowJo software (Tree Star). 2 ImmunoResearch) in 100 ␮l PBS with 2% FCS for the indicated times at Isolation of cell lines 37°C. Cells were cytospun onto poly-L-lysine-coated glass slides at 500 rpm for 3 min, fixed in acetone and methanol, and washed with PBS. Cells Embryos at embryonic day 13.5 were isolated, minced, and treated with were visualized using a Zeiss LSM510 inverted laser scanning microscope 2.5% trypsin-EDTA (Invitrogen). Cell suspensions were washed and cul- (ϫ40 original magnification, 1.3 numerical aperture objective). Images The Journal of Immunology 479

Table I. Splenic B cellsa

B cells T1 T2 Mature FoB MZ

Percentage Percentage Percentage Percentage Percentage (%) No. (%) No. (%) No. (%) No. (%) No.

Control (n ϭ 8) 60.5 Ϯ 8.0 16.6 Ϯ 7.4 15.1 Ϯ 2.3 2.5 Ϯ 1.2 22.1 Ϯ 5.0 3.7 Ϯ 1.9 31.1 Ϯ 10.7 5.1 Ϯ 2.6 4.5 Ϯ 0.8 2.5 Ϯ 1.2 Pfdn1 null (n ϭ 9) 32.5 Ϯ 12.2 1.1 Ϯ 1.5 5.8 Ϯ 5.0 0.065 Ϯ 0.108 19.9 Ϯ 4.3 0.271 Ϯ 0.409 42.1 Ϯ 10.7 0.454 Ϯ 0.673 15.0 Ϯ 3.9 0.131 Ϯ 0.152

a Number of cells given in millions. FoB, Follicular B cell. were acquired using argon and HeNe2 lasers, and the extent of capping in subsequently determined by sequencing the region of genomic ϩ IgM B cells (50 to 200) was determined visually and scored in a blind DNA adjacent to the targeting vector cloned by inverse PCR manner by an experienced microscopist. A B cell was scored capped if the (Fig. 1C). BCR polarized on half or less than half of the circumference of the cell. Only IgMϩ cells were counted as B cells in the splenocytes used in im- ES cells with the Pfdn1 mutation produced germline chimeras aging experiments. after injection into C57BL/6 blastocysts. Mice containing the mu- tated allele were genotyped by PCR using a mixture of primers Results complementary to the ﬂanking genomic DNA sequences and to the Gene trap disruption of the Pfdn1 gene LTR (Fig. 1D). Pfdn1 expression was completely disrupted in ho-

A mutation in the Pfdn1 gene (Fig. 1A) was identified in a mozygous mutant MEFs as assessed by Western blot analysis. Downloaded from screen of ES cell clones mutagenized by the GTR1.3 gene trap Thus, the 20-kDa Pfdn1 protein detected in wild-type and het- vector (23). The disrupted gene was identified by sequencing erozygous MEFs was not detected in homozygous mutant MEFs Neo-fusion transcripts amplified by 3Ј RACE, which, in the (Fig. 1E). In addition, homozygous mutant MEFs did not express case of the Pfdn1 mutation, spliced to the fourth and last exon wild-type Pfdn1 transcripts but did express lower levels of a larger of the gene (Fig. 1B), thus deleting the coiled-coil domain re- RNA fragment, presumably generated by splicing of the upstream

quired for interactions between Prefoldin and peptide substrates exons to the 3Ј Puro-LacZ sequence carried by the provirus (Fig. http://www.jimmunol.org/ (7). The exact location of the provirus in the third intron was 1F). Although these larger transcripts have the potential to express by guest on October 1, 2021

FIGURE 4. FACS analysis of splenic and peritoneal lymphocytes. A, B and T cell subpopulations. Spleens from mutant (Pfdn1Ϫ/Ϫ) mice contained proportionally more T cells (top two left panels) favoring CD4ϩ cells (bottom two left panels) than spleens from wild-type (wt) mice. B and C, B cell subpopulations. Prefoldin 1 deﬁciency preferentially reduced the proportion of immature transitional 1 (T1; AA4.1high) B cells, whereas the proportion of mature follicular B cells (FoB; CD19ϩIgMlow/IgDhigh)(top panels in B) and MZ B cells (CD19ϩIgMhighCD21high/CD23Ϫ; bottom panels in B and C) were signiﬁcantly increased. D, FACS analysis of peritoneal B cell subpopulations. The peritoneal cavities of Pfdn1Ϫ/Ϫ mice contained fewer conventional B2 as well as B1 B cells. The data shown are representative of more than three separate experiments. 480 PREFOLDIN FUNCTION IN LYMPHOCYTES

FIGURE 5. FACS analysis of thymic and BM lymphocytes. A, Prefoldin 1 deficiency reduces the proportion of double-positive (CD4ϩCD8ϩ) immature T cells and increases the double-negative (DN) immature T cells in the thymus. The developmental block appears to occur beyond the DN1 stage as Downloaded from analyzed by CD44/CD25 profile of cells within the DN lymphocyte gate (bottom panels). wt, Wild type. B, FACS analysis of BM lymphocytes. IgM/B220 and CD25/B220 profiles (top and middle panels, respectively) demonstrate relative losses of B cell precursors and increases in the proportion of pro-B cells (CD25low/B220) at the expense of pre-B cells http://www.jimmunol.org/ (CD25high/B220) (bottom panels). The developmental block appeared to occur at the earliest stage of B cell development as assessed by CD24/BP-1 profiles. C, The BM of mutant mice displayed higher proportions of fraction A (Fra A) cells (CD24low/ BP-1low) and lower proportions of fraction C and CЈ cells (CD24low/BP-1high). The data shown are representative of more than three separate experiments. by guest on October 1, 2021

aberrant Pfdn1 proteins, none was detected by Western blot anal- physical wasting. In addition, ϳ40% of the homozygous mutant ysis. Thus, the gene entrapment appeared to produce a null mice were observed with uncoordinated movement, indicating mutation. neuromuscular dysfunction.

Viability and growth of Pfdn1Ϫ/Ϫ mice Histological analysis of Pfdn1Ϫ/Ϫ mice Pfdn1Ϫ/Ϫ mice, although produced at a Mendelian ratio, were All of the homozygous mutant mice examined from 3 days to 5 wk smaller in size than wild-type and heterozygous littermates (Fig. in age were runted and displayed abnormalities in the brain and 2A), grew more slowly (Fig. 2B), and most died before 5 wk of age spleen. However, skeletal muscles, eyes, teeth, livers, kidneys, gut, (Fig. 2C). Some homozygous mutants were visibly smaller at birth. hearts, and lungs of the mutants appeared normal. Brain abnor- The average weights of mutant mice at 2, 3, and 4 wk of age were malities included hydrocephaly and neuronal loss affecting major 3.9 Ϯ 0.6g(n ϭ 14), 5.3 Ϯ 1.0g(n ϭ 13), and 4.9 Ϯ 1.0g(n ϭ commissures of the cerebrum and cerebellum. Although these 10), respectively, while the wild-type and heterozygous littermates changes, together with the hydrocephaly, made the cerebrums of weighed 8.3 Ϯ 0.8g(n ϭ 26), 12.7 Ϯ 1.8g(n ϭ 22), and 15.7 Ϯ mutant animals fragile and difﬁcult to section, the corpus callosum 1.7g(n ϭ 17), respectively. Pfdn1Ϫ/Ϫ mice lost an average of in 2- and 4-wk old mutant mice were disorganized and hypocel- 0.6 Ϯ 0.2g(n ϭ 5) during the week before death, indicating lular (Fig. 3, A and B). The arbor vitae (Fig. 3, C and D) and The Journal of Immunology 481

Table II. Thymic T cells in body weight (31.8 Ϯ 6.4% of normal). The decreased cellularity disproportionately involved CD4ϩCD8ϩ double-positive (imma- Ϫ Ϫ CD4ϪCD8Ϫ CD4ϩCD8ϩ ture) thymocytes (10% in Pfdn1 / vs 80% in wild type; Fig. 5A (%) (%) CD4ϩ (%) CD8ϩ (%) and Table II). The accumulation of CD4ϪCD8Ϫ double-negative Ϫ/Ϫ Control (n ϭ 5) 2.7 Ϯ 0.7 77.7 Ϯ 5.5 14.6 Ϯ 4.0 4.9 Ϯ 1.1 cells (5% in Pfdn1 vs 2–3% in wild type; Table II) in most Pfdn1 null (n ϭ 5) 5.2 Ϯ 3.8 10.9 Ϯ 3.1 65.5 Ϯ 2.9 18.2 Ϯ 4.5 Pfdn1-deficient mice suggests a maturation defect in the transition from double-negative to double-positive T cells. By contrast, an increase in thymic CD4ϩ and CD8ϩ mature T cells and the pres- cerebellar commissures (data not shown) were also severely dis- ence of mature TCR␤ϩ T cells in the spleen suggests that T cell organized and hypocellular. development beyond the CD4ϩCD8ϩ stage is largely unaffected. Ϫ Ϫ Blood cell counts were altered in Pfdn1 / mice, which had Consistent with a reduction in the B cell populations in the fewer numbers of both red (0.68 Ϯ 0.1 vs 1.0 Ϯ 0.03 ϫ 109 spleen and peritoneal cavity, the total number of B220ϩ cells in the cells/ml) and white cells (0.25 Ϯ 0.15 vs 0.65 Ϯ 0.1 ϫ 106 cells/ mutant BM (2.8 ϫ 104 cells per gram of body weight) was sig- ml), but hemoglobin levels appeared normal. Micronuclei were nificantly lower than in control animals (0.99 ϫ 106 cells per prominent in ϳ2% of the mutant erythrocytes as compared with gram), and the greatest reductions were seen in the proportion of Ͻ0.1% of the RBC in wild-type and heterozygous littermates (data B220ϩCD25ϩ pre-B cells, which ranged from 20 to 30% in mu- not shown). Neutrophils from mutant mice also contained more tant mice compared with 60–80% in control littermates (Fig. 5B extensively segmented nuclei (data not shown). Finally, the si- and Table III). Closer examination of the pro-B population in the nuses of mutant mice were filled with mucus and ϳ40% of the BM by the Hardy fractionation scheme (27) revealed a develop- mice showed signs of massive infection with cellular infiltrates and mental block at the earliest stage of B cell development, producing Downloaded from damage to the surrounding epithelium (Fig. 3, E and F). a large increase in the proportion of fraction A or pre-pro-B cells Spleens from Pfdn1-deficient mice were significantly smaller (Fig. 5C). Thus, the production of B cell precursors in the BM is than normal, weighing 83 Ϯ 4.4% less than the spleens from lit- severely impaired in Pfdn1-deficient mice. termate controls. Moreover, the histology of mutant spleens was abnormal in that lymphoid follicles were not clearly defined and Ϫ/Ϫ the surrounding red pulp was disorganized (Fig. 3, G and H). Total Capping defects in Pfdn1 B lymphocytes http://www.jimmunol.org/ cell numbers were reduced to 12 Ϯ 7.1% of normal in 16–25 day The effect of Pfdn1 status on signal-induced cytoskeletal reorga- old mice (data not shown), and the number of B cells was similarly nization was analyzed in B cells by a capping assay. Spleen cells reduced (Table I). The differences in spleen size and cellularity from null mutant and heterozygous control mice were stained with Ј were significant even after factoring in the sizes of mutant animals, fluorescent anti-mouse IgM F(ab )2 Abs. Control B cells displayed which were 36 Ϯ 12% smaller than normal. the expected progression from cell circumference staining or local patches to a polarized BCR or cap within 5 min, and the caps could Lymphopoiesis defects be visualized for up to 30 min (Fig. 6). By contrast, the fluorescent The proportion of splenic B cells was reduced (ϳ2–3-fold) in mu- circumference in mutant cells coalesced poorly into caps at any of by guest on October 1, 2021 tant mice with a corresponding increase in the proportion of T cells the time points analyzed (Fig. 6). These results suggest that dy- (Fig. 4A, upper panels). Specifically, only 14–33% of the spleno- namic reorganization of the cytoskeleton that accompanies the cytes from nine mutant mice examined were B220 positive as cross-linking of surface IgM is impaired in Pfdn1-deficient B cells. compared with 50–80% for wild-type or heterozygous littermates ϩ (Fig. 4A, upper panels, and Table I), and CD4 T cells were fa- Ϫ/Ϫ vored at the expense of CD8ϩ T cells (Fig. 4A, lower panels; Activation defects in Pfdn1 lymphocytes CD4:CD8 ratio 8:1; Pfdn1Ϫ/Ϫ:wild type, 2:1). Analysis of B220ϩ Reorganization and/or aggregation of surface Ag receptors is an splenic B cells revealed disproportionate reductions in immature B important component of T and B cell activation by specific Ags. cells, especially of the T1 type (IgMhighIgDlowAA4.1high; Fig. 4B We therefore asked whether the capping defects of Pfdn1-deficient and Table I), modest increases in follicular B cells, and large in- B cells and actin polymerization defects in Pfdn1-deficient T cells creases in marginal zone (MZ) B cells (Fig. 4, B and C, and Table (data not shown) were accompanied by activation defects follow- I). Although MZ B cells were significantly increased in the spleen, ing Ag receptor cross-linking. As shown in Fig. 7A, CD19ϩIgDϩ the peritoneal cavity was virtually devoid of both B1 and B2 B splenic B cells from wild-type mice were activated by either anti- cells (Fig. 4D). Together, these results suggests that the generation IgM or PMA/ionomycin as assessed by expression of the activa- of T and B cell precursors, which occurs primarily in the BM, tion marker CD86. By contrast, Pfdn1-deficient B cells were only and/or their migration to the spleen was affected to a greater extent weakly activated by anti-IgM (Fig. 7B) despite the fact that the than their maturation in the spleen. mutant cells expressed similar levels of cell surface IgM (Fig. 2D) Thymuses from mutant mice were smaller by weight and cell but were activated by PMA/ionomycin. Although less pronounced, number (9.3 Ϯ 2.7% and 4.3 Ϯ 2.3% of normal, respectively) but Pfdn1-deficient T cells also displayed activation defects following were histologically normal in appearance (data not shown). As treatment with anti-CD3 as assessed by CD69 expression (Fig. 7, before, the size reductions were greater than the overall reduction C and D).

Table III. Bone marrow B cells

B220ϩ CD25Ϫ CD25ϩ B220ϩIgMϩ B220ϩIgMhigh B220highIgMϩ Mature B220ϩ (%) IgM-Pre-Pro-B (%) Pro-B (%)a Pre-B (%)a Immature B (%) Transitional B (%) Recirculating B (%)

Control (n ϭ 7) 56.6 Ϯ 6.7 35.1 Ϯ 5.1 28.2 Ϯ 5.2 69.1 Ϯ 4.1 6.4 Ϯ 1.0 7.4 Ϯ 1.6 2.5 Ϯ 1.2 Pfdn1 null (n ϭ 8) 11.7 Ϯ 6.1 7.2 Ϯ 5.2 65.3 Ϯ 11.8 21.3 Ϯ 13.3 0.786 Ϯ 0.537 1.1 Ϯ 0.6 0.789 Ϯ 0.629

a Calculated from the B220ϩIgMϪ pre-pro-B gate. 482 PREFOLDIN FUNCTION IN LYMPHOCYTES

FIGURE 6. Capping defect in Pfdn1-deficient B cells. A, Spleno- cytes from heterozygous control and Pfdn1-deficient mice were incubated with Cy5-fluorescent labeled anti- Ј IgM F(ab )2 Abs on ice and either left on the ice (time 0) or incubated at 37°C for the indicated times. Cells

were cytospun, fixed, and analyzed as Downloaded from described in Materials and Methods. Only surface IgMϩ B cells were analyzed. Polarization of cell surface IgM into a concentrated cap was notice- ably slower and did not form a concentrated cap in Pfdn1-deficient B cells. Phase contrast image is shown http://www.jimmunol.org/ under each sample to display all cells in the field of analysis. B, The frequency of B cells with capped BCR from two independent experiments are shown. At least 50 cells were scored at each time point. by guest on October 1, 2021

Discussion mucus accumulation/infection are classic manifestations of immo- The present study describes the first genetic analysis of mamma- tile cilia syndrome (ciliary dyskinesia) caused by microtubule de- lian Prefoldin, a conserved chaperone required for efficient folding fects (28). This suggests that cytoskeletal proteins comprise the of cytoskeletal subunit proteins. Pfdn1-deficient mice developed to major substrates for Prefoldin function in mammals as in S. cer- term but displayed severe abnormalities in lymphocyte develop- evisiae. However, the phenotypes of Pfdn1-deficient mice were ment and function, mucus clearance defects, hydrocephaly, and more restricted than might be expected, considering the ubiquitous loss of nerve bundles in the CNS. Thus, mammalian Pfdn1, like the involvement of the cytoskeleton in diverse cellular processes (2, orthologous genes of Saccharomyces cerevisiae and C. elegans, 3). Apparently, basic functions of the cytoskeleton in processes encodes nonredundant functions that cannot be replaced by other such as cell division, cell shape, and intracellular trafficking were chaperones. The requirement for Pfdn1 is also consistent with bio- not sufficiently impaired to affect the development of most tissues, chemical reconstitution experiments in which all six subunits are at least as viewed on an anatomical level. For example, Prefoldin- required for assembly and function of the Prefoldin complex (9, deficient mice lacked neural tube closure defects often associated 10, 13). with genes involved in actin regulation and mitotic function (29). Pfdn1 mutant phenotypes, although varied, were all consistent Because yeast Prefoldin enhances the rate of actin folding with with abnormal cytoskeletal function. The defects in B and T cell little effect on steady-state protein levels (13), cellular processes development and function were similar to but more severe than that require rapid and extensive cytoskeletal synthesis or remod- mutations in proteins (WASP and WIP) that regulate actin poly- eling, such as immune cell signaling and axon guidance, may be merization in lymphocytes (5). Actin and tubulin cytoskeletons are more sensitive to the loss of Pfdn1 function. prominent features of the neural growth cone and orchestrate axon Commissures consist of major tracts of neurons that connect outgrowth and guidance (4). Finally, hydrocephaly and respiratory regions of the CNS. Commissure defects frequently result from The Journal of Immunology 483

with WASP in resting T cells, and the complex translocates to the immune synapse following TCR engagement. WIP-deficient T cells develop normally but display cell activation and actin assembly defects after TCR ligation and conjugate less efficiently with super Ag-presenting B cells (39). Cell activation and Ag receptor capping defects in Pfdn1-deficient T and B cells recapitulate features of the WASP and WIP knockouts. However, the effects of the Pfdn1 mutation on lymphocyte development and function were more severe, as evidenced by developmental and functional defects affecting both T and B cells. The phenotype of Pfdn1-deficient mice is consistent with an actin-folding defect that could indirectly influence cytoskeleton assembly at a level upstream of WASP and WIP, although the overall consequences of Pfdn1 mutation were more restricted than those involving other regulators of actin assembly, including N- WASP (40, 41), Scar/WAVE2 (42, 43), and Nck1 together with Nck2 (44), because the latter impair embryonic development. Lymphopoiesis defects in Pfdn1Ϫ/Ϫ mice included dramatic re- ϩ ϩ

ductions in pre-B cells in the BM and in immature CD4 CD8 Downloaded from double-positive T cells in the thymus. The eventual production of mature B and T cells suggests that Prefoldin is required primarily FIGURE 7. Activation defect in Pfdn1-deﬁcient B (A and B) and T cells during the genesis and/or renewal of lymphocyte precursors rather (C and D). CD19ϩIgMϩIgDϩ spleen B cells from wild-type (A) and Pfdn1- Ј than at later stages of lymphocyte development. These develop- deﬁcient mice (B) were treated with anti-IgM F(ab )2 (dark lines), PMA/ ionomycin (dotted lines), or medium alone (gray lines), and after 16 h cell mental defects precede the expression of functional Ag receptors

activation was assessed by flow cytometry. Pfdn1-deficient B cells were and therefore differ from the activation defects observed in mature http://www.jimmunol.org/ activated by PMA/ionomycin but not by anti-IgM as assessed by anti- B and T cells. However, the defects in both B and T cell devel- CD86 staining. The mean fluorescent intensities (MFIs) of wild-type and opment occurred at stages that require cell migration and physical mutant B cells before and after anti-IgM treatment were 95 and 339, interactions with stromal cells in the microenvironment of the thy- ϩ whereas in mutant B cells the MFIs were 48 and 66, respectively. CD4 T mus and BM (45, 46). An additional developmental defect result- cells from wild-type (C) and Pfdn1-deficient mice (D) were treated with ing from impaired BCR signaling is suggested by substantial in- anti-CD3 (dark lines), PMA/ionomycin (dotted lines), or medium alone creases in the proportion of MZ B cells relative to follicular B cells (gray lines), and after 16 h cell activation was assessed by flow cytometry. Pfdn1-deficient T cells were activated by PMA/ionomycin but not by anti- in Pfdn1 mutant mice, because the formation of MZ B cells is CD3 as assessed by anti-CD69 staining. The MFIs (CD69) of wild-type/ thought to require a lower threshold of Ag receptor signaling than mutant T cells before and after anti-CD3 treatment were 6 and 151, mature follicular B cells (47). Alternatively, immune cell defects by guest on October 1, 2021 whereas MFI of mutant T cells were 5 and 28, respectively. could be a secondary consequence of cellular or systemic stress in the mutant animals. Additional experiments will be required to clarify the roles of Prefoldin in early lymphocyte development and mutations in genes such as Enah (Mena) (30, 31), Mapk8 (Jnk1) Ag receptor-dependent lymphocyte maturation. (32) Nr2f1 (COUP-TFI) (33), Mtap1b (MAP1B) (34), and possibly Lymphocyte activation and actin assembly defects raise ques- Scar/WAVE1 (35, 36), which participate in neuron outgrowth/ tions about the mechanism by which Pfdn1 mutation influences axon guidance by influencing cytoskeletal expression and function. cytoskeletal function. The fact that Prefoldin (Gim) function is not These phenotypes illustrate the importance of the actin and tubulin essential for cell viability in S. cerevisiae (9, 10) has been attrib- cytoskeletons in the neural growth cone, which is present at the uted to the ability of substrates to fold at slower rates without distal end of extending axons and orchestrates axon outgrowth and functionally interacting with Prefoldin (13, 48). However, if Pre- guidance (4). Similarly, the CNS phenotypes in Pfdn1-deficient foldin simply enhances the rate of protein folding, then resting mice are consistent with defects in actin and/or tubulin folding. lymphocytes should contain adequate pools of properly folded pro- Chaperone mutations can also cause neurodegenerative diseases teins for processes, such as F-actin assembly and receptor capping, due to the toxic effects of misfolded proteins (20, 22). Although that occur within minutes after Ag receptor cross-linking. Indeed, neuronal phenotypes in Pfdn1-deficient mice could result from wild-type and Pfdn1-deficient lymphocytes express similar levels toxicity caused by aggregates of misfolded proteins, this seems of ␤-actin protein as assessed by Western blot analysis (data not less likely considering the overall pattern of neuronal cell loss. shown), suggesting that the activation and actin assembly defects Involvement of the actin cytoskeleton in lymphocyte function is does not result from inadequate pools of cytoskeletal subunit pro- illustrated by mutations in the genes encoding WASP (37, 38) and teins. In principle, Prefoldin could play other roles in cytoskeletal the WASP-interacting protein WIP (39). WASP belongs to a fam- assembly and/or remodeling that are separate from cotranslational ily of proteins (including N-WASP and Scar/WAVEs) that activate protein folding. In addition, Pfdn1-deficient cells could harbor suf- the actin-nucleating activity of the Arp2/3 (actin-related protein 2 ficient levels of improperly folded subunits to poison polymer as- and 3) complex. WASP, when recruited to an activated TCR, in- sembly upon cell activation. Although additional experiments will teracts with the Arp2/3 complex to promote cytoskeletal reorga- be required to examine these issues, the effects of Pfdn1 deficiency nization. Rapid actin polymerization provides a motile force that resemble aging-associated defects in lymphocyte development and physically clusters TCRs into a synapse and generates a scaffold of function (49, 50). Considering that altered cellular proteins accu- actin filaments involved in sustained cell signaling (reviewed in mulate during aging and contribute to a number of chronic diseases Ref. 5). WASP expression is restricted to hematopoietic cells, thus (21, 22), the present study implicates misfolded cytoskeletal pro- limiting the phenotype of WASP knockout mice. WIP interacts teins as a cause of age-related defects in lymphocyte function. 484 PREFOLDIN FUNCTION IN LYMPHOCYTES

Acknowledgments developmental stage-specific regulation of diacylglycerol and calcium signaling upon B cell receptor engagement. J. Immunol. 177: 5405–5413. We thank Abudi Nashabi for technical assistance, Drs. Danyvid Olivares- 26. Bezbradica, J. S., L. E. Gordy, A. K. Stanic, S. Dragovic, T. Hill, T. Hawiger, Villagomez and Luc Van Kaer for help with FACS, Dr. Robert Collins for D. Unumatz, L. Van Kaer, and S. Joyce. 2006. Granulocyte-macrophage colony- histopathology analysis, and Dr. Emily Clark for expert analysis of BCR stimulating factor regulates effector differentiation of invariant natural killer T capped B cells. cells during thymic ontogeny. Immunity 25: 487–497. 27. Hardy, R. R., C. E. Carmack, S. A. Shinton, J. D. Kemp, and K. Hayakawa. 1991. Resolution and characterization of pro-B and pre-pro-B cell stages in normal Disclosures mouse bone marrow. J. Exp. Med. 173: 1213–1225. The authors have no financial conflict of interest. 28. Afzelius, B. A. 2004. Cilia-related diseases. J. Pathol. 204: 470–477. 29. Juriloff, D. M., and M. J. Harris. 2000. Mouse models for neural tube closure defects. Hum. Mol. Genet. 9: 993–1000. References 30. Menzies, A. S., A. Aszodi, S. E. Williams, A. Pfeifer, A. M. Wehman, K. L. Goh, 1. Small, J. V., K. Rottner, and I. Kaverina. 1999. Functional design in the actin C. A. Mason, R. Fassler, and F. B. Gertler. 2004. Mena and vasodilator-stimu- cytoskeleton. Curr. Opin. Cell Biol. 11: 54–60. lated phosphoprotein are required for multiple actin-dependent processes that 2. Revenu, C., R. Athman, S. Robine, and D. Louvard. 2004. The co-workers of shape the vertebrate nervous system. J. Neurosci. 24: 8029–8038. actin filaments: from cell structures to signals. Nat. Rev. Mol. Cell Biol. 5: 31. Lanier, L. M., M. A. Gates, W. Witke, A. S. Menzies, A. M. Wehman, 635–646. J. D. Macklis, D. Kwiatkowski, P. Soriano, and F. B. Gertler. 1999. Mena is 3. Galjart, N. 2005. CLIPs and CLASPs and cellular dynamics. Nat. Rev. Mol. Cell required for neurulation and commissure formation. Neuron 22: 313–325. Biol. 6: 487–498. 32. Chang, L., Y. Jones, M. H. Ellisman, L. S. Goldstein, and M. Karin. 2003. JNK1 4. Dent, E. W., and F. B. Gertler. 2003. Cytoskeletal dynamics and transport in is required for maintenance of neuronal microtubules and controls phosphoryla- growth cone motility and axon guidance. Neuron 40: 209–227. tion of microtubule-associated proteins. Dev. Cell 4: 521–533. 5. Dustin, M. L., and J. A. Cooper. 2000. The immunological synapse and the actin 33. Armentano, M., A. Filosa, G. Andolfi, and M. Studer. 2006. COUP-TFI is re- cytoskeleton: molecular hardware for T cell signaling. Nat. Immunol. 1: 23–29. quired for the formation of commissural projections in the forebrain by regulating 6. Young, J. C., V. R. Agashe, K. Siegers, and F. U. Hartl. 2004. Pathways of axonal growth. Development 133: 4151–4162. 34. Takei, Y., J. Teng, A. Harada, and N. Hirokawa. 2000. Defects in axonal elon-

chaperone-mediated protein folding in the cytosol. Nat. Rev. Mol. Cell Biol. 5: Downloaded from 781–791. gation and neuronal migration in mice with disrupted tau and map1b genes. 7. Siegert, R., M. R. Leroux, C. Scheufler, F. U. Hartl, and I. Moarefi. 2000. Struc- J. Cell Biol. 150: 989–1000. ture of the molecular chaperone prefoldin: unique interaction of multiple coiled 35. Dahl, J. P., J. Wang-Dunlop, C. Gonzales, M. E. Goad, R. J. Mark, and coil tentacles with unfolded proteins. Cell 103: 621–632. S. P. Kwak. 2003. Characterization of the WAVE1 knock-out mouse: implica- 8. Martin-Benito, J., J. Boskovic, P. Gomez-Puertas, J. L. Carrascosa, C. T. Simons, tions for CNS development. J. Neurosci. 23: 3343–3352. S. A. Lewis, F. Bartolini, N. J. Cowan, and J. M. Valpuesta. 2002. Structure of 36. Soderling, S. H., L. K. Langeberg, J. A. Soderling, S. M. Davee, R. Simerly, eukaryotic prefoldin and of its complexes with unfolded actin and the cytosolic J. Raber, and J. D. Scott. 2003. Loss of WAVE-1 causes sensorimotor retardation chaperonin CCT. EMBO J. 21: 6377–6386. and reduced learning and memory in mice. Proc. Natl. Acad. Sci. USA 100: 9. Geissler, S., K. Siegers, and E. Schiebel. 1998. A novel protein complex pro- 1723–1728. http://www.jimmunol.org/ moting formation of functional ␣- and ␥-tubulin. EMBO J. 17: 952–966. 37. Acuto, O., and D. Cantrell. 2000. T cell activation and the cytoskeleton. Annu. 10. Vainberg, I. E., S. A. Lewis, H. Rommelaere, C. Ampe, J. Vandekerckhove, Rev. Immunol. 18: 165–184. H. L. Klein, and N. J. Cowan. 1998. Prefoldin, a chaperone that delivers unfolded 38. Snapper, S. B., F. S. Rosen, E. Mizoguchi, P. Cohen, W. Khan, C. H. Liu, proteins to cytosolic chaperonin. Cell 93: 863–873. T. L. Hagemann, S. P. Kwan, R. Ferrini, L. Davidson, et al. 1998. Wiskott- 11. Spiess, C., A. S. Meyer, S. Reissmann, and J. Frydman. 2004. Mechanism of the Aldrich syndrome protein-deficient mice reveal a role for WASP in T but not B eukaryotic chaperonin: protein folding in the chamber of secrets. Trends Cell cell activation. Immunity 9: 81–91. Biol. 14: 598–604. 39. Anton, I. M., M. A. de la Fuente, T. N. Sims, S. Freeman, N. Ramesh, 12. Hansen, W. J., N. J. Cowan, and W. J. Welch. 1999. Prefoldin-nascent chain J. H. Hartwig, M. L. Dustin, and R. S. Geha. 2002. WIP deficiency reveals a complexes in the folding of cytoskeletal proteins. J. Cell Biol. 145: 265–277. differential role for WIP and the actin cytoskeleton in T and B cell activation. 13. Siegers, K., T. Waldmann, M. R. Leroux, K. Grein, A. Shevchenko, E. Schiebel, Immunity 16: 193–204. and F. U. Hartl. 1999. Compartmentation of protein folding in vivo: sequestration 40. Lommel, S., S. Benesch, K. Rottner, T. Franz, J. Wehland, and R. Kuhn. 2001. of non-native polypeptide by the chaperonin-GimC system. EMBO J. 18: 75–84. Actin pedestal formation by enteropathogenic Escherichia coli and intracellular by guest on October 1, 2021 14. Le Bot, N., M. C. Tsai, R. K. Andrews, and J. Ahringer. 2003. TAC-1, a regulator motility of Shigella flexneri are abolished in N-WASP-defective cells. EMBO of microtubule length in the C. elegans embryo. Curr. Biol. 13: 1499–1505. Rep. 2: 850–857. 15. Zipperlen, P., A. G. Fraser, R. S. Kamath, M. Martinez-Campos, and J. Ahringer. 41. Snapper, S. B., F. Takeshima, I. Anton, C. H. Liu, S. M. Thomas, D. Nguyen, 2001. Roles for 147 embryonic lethal genes on C. elegans chromosome I iden- D. Dudley, H. Fraser, D. Purich, M. Lopez-Ilasaca, et al. 2001. N-WASP defi- tified by RNA interference and video microscopy. EMBO J. 20: 3984–3992. ciency reveals distinct pathways for cell surface projections and microbial actin- 16. Siegers, K., B. Bolter, J. P. Schwarz, U. M. Bottcher, S. Guha, and F. U. Hartl. based motility. Nat. Cell Biol. 3: 897–904. 2003. TRiC/CCT cooperates with different upstream chaperones in the folding of 42. Yan, C., N. Martinez-Quiles, S. Eden, T. Shibata, F. Takeshima, R. Shinkura, distinct protein classes. EMBO J. 22: 5230–5240. Y. Fujiwara, R. Bronson, S. B. Snapper, M. W. Kirschner, et al. 2003. WAVE2 17. Ho, J. C., K. N. Chan, W. H. Hu, W. K. Lam, L. Zheng, G. L. Tipoe, J. Sun, deficiency reveals distinct roles in embryogenesis and Rac-mediated actin-based R. Leung, and K. W. Tsang. 2001. The effect of aging on nasal mucociliary motility. EMBO J. 22: 3602–3612. clearance, beat frequency, and ultrastructure of respiratory cilia. Am. J. Respir. 43. Yamazaki, D., S. Suetsugu, H. Miki, Y. Kataoka, S. Nishikawa, T. Fujiwara, Crit. Care Med. 163: 983–988. N. Yoshida, and T. Takenawa. 2003. WAVE2 is required for directed cell mi- 18. Svartengren, M., R. Falk, and K. Philipson. 2005. Long-term clearance from gration and cardiovascular development. Nature 424: 452–456. small airways decreases with age. Eur. Respir. J. 26: 609–615. 44. Bladt, F., E. Aippersbach, S. Gelkop, G. A. Strasser, P. Nash, A. Tafuri, 19. Ramaekers, F. C., and F. T. Bosman. 2004. The cytoskeleton and disease. F. B. Gertler, and T. Pawson. 2003. The murine Nck SH2/SH3 adaptors are J. Pathol. 204: 351–354. important for the development of mesoderm-derived embryonic structures and for 20. Barral, J. M., S. A. Broadley, G. Schaffar, and F. U. Hartl. 2004. Roles of mo- regulating the cellular actin network. Mol. Cell. Biol. 23: 4586–4597. lecular chaperones in protein misfolding diseases. Semin. Cell Dev. Biol. 15: 45. Nagasawa, T. 2006. Microenvironmental niches in the bone marrow required for 17–29. B-cell development. Nat. Rev. Immunol. 6: 107–116. 21. Hipkiss, A. R. 2006. Accumulation of altered proteins and ageing: causes and 46. Takahama, Y. 2006. Journey through the thymus: stromal guides for T-cell de- effects. Exp. Gerontol. 41: 464–473. velopment and selection. Nat. Rev. Immunol. 6: 127–135. 22. Macario, A. J., and E. Conway de Macario. 2005. Sick chaperones, cellular stress, 47. Pillai, S., A. Cariappa, and S. T. Moran. 2005. Marginal zone B cells. Annu. Rev. and disease. N. Engl. J. Med. 353: 1489–1501. Immunol. 23: 161–196. 23. Lin, Q., S. L. Donahue, T. Moore-Jarrett, S. Cao, A. B. Osipovich, and 48. Neirynck, K., D. Waterschoot, J. Vandekerckhove, C. Ampe, and H. E. Ruley. 2006. Mutagenesis of diploid mammalian genes by gene entrapment. H. Rommelaere. 2005. Actin interacts with CCT via discrete binding sites: a Nucleic Acids Res. 34: e139. binding transition-release model for CCT-mediated actin folding. J. Mol. Biol. 24. Williamson, D. J., S. Banik-Maiti, J. DeGregori, and H. E. Ruley. 2000. hnRNP 355: 124–138. C is required for postimplantation mouse development but is dispensable for cell 49. Linton, P. J., and K. Dorshkind. 2004. Age-related changes in lymphocyte de- viability. Mol. Cell. Biol. 20: 4094–4105. velopment and function. Nat. Immunol. 5: 133–139. 25. Hoek, K. L., P. Antony, J. Lowe, N. Shinners, B. Sarmah, S. R. Wente, D. Wang, 50. Sadighi Akha, A. A., and R. A. Miller. 2005. Signal transduction in the aging R. M. Gerstein, and W. N. Khan. 2006. Transitional B cell fate is associated with immune system. Curr. Opin. Immunol. 17: 486–491.