An Extended Nomenclature for Mammalian V-ATPase Subunit Genes and Splice Variants
Kevin C. Miranda1,2, Fiona E. Karet3*, Dennis Brown1,2*
1 Program in Membrane Biology and Division of Nephrology, Center for Systems Biology, Massachusetts General Hospital, Boston, Massachusetts, United States of America, 2 Harvard Medical School, Boston, Massachusetts, United States of America, 3 Department of Medical Genetics, University of Cambridge, Cambridge, United Kingdom

Abstract
The vacuolar-type H+-ATPase (V-ATPase) is a multisubunit proton pump that is involved in both intra- and extracellular acidification processes throughout the body. Multiple homologs and splice variants of V-ATPase subunits are thought to explain its varied spatial and temporal expression pattern in different cell types. Recently subunit nomenclature was standardized with a total of 22 subunit variants identified. However this standardization did not accommodate the existence of splice variants and is therefore incomplete. Thus, we propose here an extension of subunit nomenclature along with a literature and sequence database scan for additional V-ATPase subunits. An additional 17 variants were pulled from a literature search while 4 uncharacterized potential subunit variants were found in sequence databases. These findings have been integrated with the current V-ATPase knowledge base to create a new V-ATPase subunit catalogue. It is envisioned this catalogue will form a new platform on which future studies into tissue- and organelle-specific V-ATPase expression, localization and function can be based.
Citation: Miranda KC, Karet FE, Brown D (2010) An Extended Nomenclature for Mammalian V-ATPase Subunit Genes and Splice Variants. PLoS ONE 5(3): e9531. doi:10.1371/journal.pone.0009531 Editor: Alan Ruttenberg, Science Commons, United States of America Received October 2, 2009; Accepted February 3, 2010; Published March 10, 2010 Copyright: ß 2010 Miranda et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: KCM is funded through a National Kidney Foundation Research Fellowship. FEK is funded through the Wellcome Trust, while DB is funded by a National Institutes of Health grant DK42956 (DB). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. * E-mail: brown@receptor.mgh.harvard.edu (DB); fek1000@cam.ac.uk (FEK)

Introduction
The vacuolar-type H -ATPase (V-ATPase) is a proton pump found in all nucleated cells of the body. Partially embedded within the membrane, its function is to transfer two hydrogen ions out of the cytoplasm at the expense of 1 ATP molecule [1]. It is thereby able to establish a proton gradient within the lumen of organelles such as lysosomes, endosomes and the trans-Golgi network. Organelle acidification is important for a diverse array of functions including intracellular trafficking and protein degradation [2–4]. The V-ATPase is also functionally important at the plasma membrane of specialized cell types in certain tissues. There, it is responsible for critical homeostatic functions such as body acidbase regulation (renal proximal tubule and collecting duct intercalated cells), bone remodeling (osteoclasts), and sperm storage and maturation (clear cells in the epididymis) [2,5,6] as well as other potential functions in other organs. Loss of V-ATPase activity due to subunit mutations in specific cell types has been implicated in diverse pathophysiological states such as kidney and bone disease, sensorineural deafness and wrinkly skin syndrome [7–10]. Thus V-ATPase functions in a very broad array of physiological processes. The V-ATPase is a large (800 kDa) and complex molecular motor. It is made up of at least 13 individual components/protein subunits organized into two functional domains: V0 and V1 [1,4,11,12]. The V0 domain is composed of several transmembrane subunits that are involved in hydrogen ion translocation across the bilayer, while the V1 domain is a peripheral to the
PLoS ONE | www.plosone.org 1
+

membrane and is the site of ATP hydrolysis. V0 is composed of 5 subunits labeled a to e while V1 consists of 8 subunits denoted A to H. All 13 different subunits are encoded by separate genes located throughout the genome. Many of the 13 V-ATPase subunits exist as homologs, thereby adding another level of complexity to the motor. The diverse functions and locations of V-ATPases are believed to be encoded within the various homologs. For example the d1 subunit is ubiquitously expressed, while the d2 homolog is seen only in the kidney, osteoclast and lung [13]. Similarly two isoforms of the B subunit were initially described as so-called ‘‘kidney’’ (B1) and ‘‘brain’’ (B2) specific isoforms, although it is now clear that expression of B1 is not restricted to the kidney [12]. Currently, homologs have been identified for the a, d, e, B, C, E and G subunits [11,14]. An additional level of V-ATPase subunit variation is encoded through splice variants. To date, splice variants have been identified for the a, d, e, C, G and H subunits [9,13–23]. Just like homologs, splice variants have been shown to exhibit different expression patterns. For example two splice variants of subunit a1 are expressed in rat neurons: one variant localized to axonal varicosities while the other was sorted to distal dendrites and axons [17]. To date, discovery of V-ATPase homologs and splice variants has largely been subunit focused and experimentally based. This method has proven successful as demonstrated by the large number of identified variants. However, this fragmented discovery process led to a fragmented naming system which was recently
March 2010 | Volume 5 | Issue 3 | e9531

V-ATPase Subunit Genes

Table 1. Overview of V-ATPase subunits and their respective RefSeq accession numbers.
Augmented Nomenclature of Human V-ATPase Subunit Genes Gene Symbol Augmented Nomenclature Nucleotide Accession Number

Table 1. Cont.

Ref

Sp

Note

V0 Membrane Sector 1 2 3 4# 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 ATP6V0A1 ---------------------ATP6V0A2 -------TCIRG1 -------ATP6V0A4 --------------ATP6V0B -------ATP6V0C ATP6V0D1 ATP6V0D2 ATP6V0E1 ATP6V0E2 --------------ATP6V0A1v1 ATP6V0A1v2 ATP6V0A1v3 ATP6V0A1v4 ATP6V0A2v1 ATP6V0A2v2 ATP6V0A3v1 ATP6V0A3v2 ATP6V0A4v1 ATP6V0A4v2 ATP6V0A4v3 ATP6V0Bv1 ATP6V0Bv2 ATP6V0C* ATP6V0D1* ATP6V0D2* ATP6V0E1* ATP6V0E2v1 ATP6V0E2v2 ATP6V0E2v3 NM_001130020 NM_001130021 NM_005177 DQ286422 NM_012463 BC068531 NM_006019 NM_006053 NM_020632 NM_130840 NM_130841 NM_004047 NM_001039457 NM_001694 NM_004691 NM_152565 NM_003945 NM_145230 AK098362 NM_001100592 [15] [16] [15] [17] [29] [43] [18] [19] [9] [9] [5] [30] [31] [32] [13] [33] [14] B M B R B H H H H H H B B H B H K,E E,L E,L L K,E D K,E E,L K,E E,L E K,E E K,E K,E K,E K,E E,L D E

subunits described in this report. An augmented naming system (Proposed) is implemented to incorporate the multiple splice variants identified in this analysis. Notes for Table 1. Ref - Original cloning paper. Sp - Species gene was originally cloned from: B - Bovine; M - Mouse; R - Rat; H - Human; C- Chicken. K - known variant noted in Smith AN et al (2003). E - entrez gene entry exists. L - identified in literature search. D - cloned in a high-throughput experiment, needs further experimental validation. #No human transcript could be identified in RefSeq, so the rat ortholog is provided. *If a novel homolog is discovered for a subunit with no known homologs then the current homolog will be denoted 21 and the novel homolog denoted 22. For example ATP6V1F will become ATP6V1F1 and the novel homolog ATP6V1F2. If additional splice variants of these new homologs are then discovered, the terminology will become, for example, ATP6V1F1v1 and ATP6V1F1v2 etc. doi:10.1371/journal.pone.0009531.t001

standardized [11]. A large effort was undertaken to associate the multiple names of a given subunit to one new standard nomenclature, but extension of this nomenclature to include splice-variants was not systematically pursued at that time. Thus, here we have performed a literature search of all V-ATPase subunits, their homologs and splice variants. In addition we scanned both the RefSeq nucleotide and protein databases [24] to identify any novel subunits that have been deposited within these databases. The results of this analysis are presented here.

Results and Discussion
At the time of the first V-ATPase nomenclature standardization, the incorporation of splice-variants was not implemented [11]. Thus, we now suggest augmenting the naming policy to allow for the differentiation of splice variants. In accordance with HUGO Gene Nomenclature Committee (HGNC) specifications, we propose the addition of a ‘‘v[1..x]’’ suffix to the relevant gene symbols, and ‘‘i[1..x]’’ to the predicted proteins. For example, ATP6V0A1v1 and ATP6V0A1v2 would differentiate between splice-variants 1 and 2 of subunit a1. As is the case at present, we propose that if new homologs of subunits with only a single currently known isoform are identified in the future, they should be named in numerical order (e. g., ATP6V1F will become ATP6V1F1 if a new isoform, which will become ATP6V1F2, is discovered). We also propose that the first full-length ‘‘known’’ subunit splice variant should be named ‘‘v1’’, allowing for subsequent addition and standard nomenclature if additional variants are discovered in the future. For subunits with only one known variant, the v1 nomenclature should be appended to the existing name only if and when a second variant (which will be named v2) is identified. In view of our extensive database search we believe, however, that this possibility is unlikely. The V-ATPase proton pump is composed of 13 subunits. According to the data presented in Smith et al (2003) these 13 are encoded through 22 homologs. These are denoted with a ‘‘K’’ for known in Table 1. A simple search of the literature and the Entrez Gene database [24] reveals another 17 variants. These are denoted with either an ‘‘E’’ for Entrez or ‘‘L’’ for literature in Table 1. All of these additional subunits are splice variants except for e2, which was cloned after the publication of Smith et al (2003). Our computational analysis of sequence repositories has identified
2 March 2010 | Volume 5 | Issue 3 | e9531

V1 Peripheral Sector 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 ATP6V1A ATP6V1B1 ATP6V1B2 ATP6V1C1 ATP6V1C2 -------ATP6V1D -------ATP6V1E1 --------------ATP6V1E2 ATP6V1F ATP6V1G1 -------ATP6V1G2 -------ATP6V1G3 --------------ATP6V1H --------------ATP6V1A* ATP6V1B1* ATP6V1B2* ATP6V1C1* ATP6V1C2v1 ATP6V1C2v2 ATP6V1Dv1* ATP6V1Dv2* ATP6V1E1v1 ATP6V1E1v2 ATP6V1E1v3 ATP6V1E2* ATP6V1F* ATP6V1G1v1 ATP6V1G1v2 ATP6V1G2v1 ATP6V1G2v2 ATP6V1G3v1 ATP6V1G3v2 ATP6V1G3v3 ATP6V1Hv1 ATP6V1Hv2 ATP6V1Hv3 NM_001690 NM_001692 NM_001693 NM_001695 NM_144583 NM_001039362 NM_015994 AF100741 NM_001696 NM_001039366 NM_001039367 NM_080653 NM_004231 NM_004888 BC008452 NM_130463 NM_138282 NM_133262 NM_133326 BC101129 NM_015941 NM_213619 NM_213620 [34] [35] [36] [37] [13] [13] [38] [44] [39] [40] [41] [42] [43] [21] [22] [13] [13] [23] [23] B H H B H H M H B H H B H,M H K,E K,E K,E K,E K,E L,E K,E D K,E E E K,E K,E K,E D K,E

R,B,C L,E H H B B K,E E L K,E L,E E

A list of the current HUGO Gene Nomenclature Committee (HGNC) designations for the known V-ATPase subunits is provided, along with the corresponding

PLoS ONE | www.plosone.org

V-ATPase Subunit Genes

another 4 potentially novel V-ATPase variants. Again, all new sequences are encoded by splice variants; these are denoted with a ‘‘D’’ for database search in Table 1. It should be noted that the additional 4 variants were cloned in high-throughput mRNA discovery studies. Thus, they have no accompanying experimental validation at the protein level; instead all exist as sequenced and apparently full length mRNA transcripts. Further experimental work will be required to verify the existence of these clones in vivo; nonetheless we have included them here. Thus, with this simple literature and database search we have increased the number of V-

ATPase subunit variants from 22 to 43 (Figure 1). Finally, we propose that accessory proteins (whose functions related to the VATPase remain unknown) that also use the ATP6 nomenclature, including ATP6AP1 and ATP6AP2, should be included in the revised nomenclature scheme, and the appropriate suffixes should be added to their names if additional isoforms and splice variants emerge in the future. However, we have also examined these sequences and apart from pseudogenes, found no variants. Experimentally identifying and characterizing the associated functional proteins was outside the scope of this study but should

Figure 1. A schematic view of V-ATPase subunits. The exonic structure of all V-ATPase subunits is shown; untranslated regions are shown in light grey while protein coding regions are black. The intronic distances have been divided by 15 for display purposes. Refer to Table 1 for further information on subunits. doi:10.1371/journal.pone.0009531.g001

PLoS ONE | www.plosone.org

3

March 2010 | Volume 5 | Issue 3 | e9531

V-ATPase Subunit Genes

be the subject of future work by investigators interested in VATPase function, including our own groups. The results of this study suggest that the V-ATPase is regulated in a much more complex manner than is currently assumed. At what level this regulation is exerted remains to be determined experimentally, and having a complete systems overview of all V-ATPase components will help expedite this process. This almost two-fold increase in the number of V-ATPase subunit forms demonstrates how the output of broad genomic scale projects can be utilized for a specialized pursuit. It also highlights the importance of computational methods for sifting and sorting through vast amounts of data deposited in sequence databases worldwide. Identification and standardization of transcript variation offers a powerful approach to guide the future assignation of functional significance among protein variants.

Materials and Methods
Sequences corresponding to the 13 human V-ATPase subunits described previously [11] were retrieved from the HUGO Gene Nomenclature Database (HGNC) database [25]. These sequences were used to query the RefSeq non-redundant protein and nucleotide databases [24] with the BLAST algorithm [26]. Various modules from the Bioperl toolkit were utilized to process the resulting BLAST output [27]. The HGNC V-ATPase identifiers were used to query the SpliceCenter databases [28]. The results of this analysis were manually integrated with the BLAST results described above. This integrated catalogue is presented in Table 1 and Figure 1.

Author Contributions
Conceived and designed the experiments: KM. Performed the experiments: KM FK. Analyzed the data: KM FK DB. Contributed reagents/ materials/analysis tools: KM FK DB. Wrote the paper: KM FK DB.

References
1. Forgac M (2007) Vacuolar ATPases: Rotary proton pumps in physiology and pathophysiology. Nature Rev Mol Cell Biol 8(11): 917–929. 2. Brown D, Breton S (2000) H+V-ATPase-dependent luminal acidification in the kidney collecting duct and the epididymis/vas deferens: Vesicle recycling and transcytotic pathways. J Exp Biol 203(Pt 1): 137–145. 3. Futai M, Oka T, Sun-Wada G, Moriyama Y, Kanazawa H, et al. (2000) Luminal acidification of diverse organelles by V-ATPase in animal cells. J Exp Biol 203(Pt 1): 107–116. 4. Nelson N, Harvey WR (1999) Vacuolar and plasma membrane protonadenosinetriphosphatases. Physiol Rev 79(2): 361–385. 5. Blair HC, Teitelbaum SL, Ghiselli R, Gluck S (1989) Osteoclastic bone resorption by a polarized vacuolar proton pump. Science 245(4920): 855–857. 6. Breton S, Smith PJ, Lui B, Brown D (1996) Acidification of the male reproductive tract by a proton pumping H+-ATPase. Nat Med 2(4): 470–472. 7. Frattini A, Orchard PJ, Sobacchi C, Giliani S, Abinun M, et al. (2000) Defects in TCIRG1 subunit of the vacuolar proton pump are responsible for a subset of human autosomal recessive osteopetrosis. Nat Genet 25(3): 343–346. 8. Karet FE, Finberg KE, Nelson RD, Nayir A, Mocan H, et al. (1999) Mutations in the gene encoding B1 subunit of H+-ATPase cause renal tubular acidosis with sensorineural deafness. Nat Genet 21(1): 84–90. 9. Smith AN, Skaug J, Choate KA, Nayir A, Bakkaloglu A, et al. (2000) Mutations in ATP6N1B, encoding a new kidney vacuolar proton pump 116-kD subunit, cause recessive distal renal tubular acidosis with preserved hearing. Nat Genet 26(1): 71–75. 10. Kornak U, Reynders E, Dimopoulou A, Reeuwijk Jv, Fischer B, et al. (2008) Impaired glycosylation and cutis laxa caused by mutations in the vesicular H+ATPase subunit ATP6V0A2. Nat Genet 40(1): 32–34. 11. Smith AN, Lovering RC, Futai M, Takeda J, Brown D, et al. (2003) Revised nomenclature for mammalian vacuolar-type H+-ATPase subunit genes. Mol Cell 12(4): 801–803. 12. Wagner CA, Finberg KE, Breton S, Marshansky V, Brown D, et al. (2004) Renal vacuolar H+-ATPase. Physiol Rev 84(4): 1263–1314. 13. Smith AN, Borthwick KJ, Karet FE (2002) Molecular cloning and characterization of novel tissue-specific isoforms of the human vacuolar H+-ATPase C, G and d subunits, and their evaluation in autosomal recessive distal renal tubular acidosis. Gene 297(1–2): 169–177. 14. Blake-Palmer KG, Su Y, Smith AN, Karet FE (2007) Molecular cloning and characterization of a novel form of the human vacuolar H+-ATPase e-subunit: An essential proton pump component. Gene 393(1–2): 94–100. 15. Peng SB, Crider BP, Xie XS, Stone DK (1994) Alternative mRNA splicing generates tissue-specific isoforms of 116-kDa polypeptide of vacuolar proton pump. J Biol Chem 269(25): 17262–17266. 16. Nishi T, Forgac M (2000) Molecular cloning and expression of three isoforms of the 100-kDa a subunit of the mouse vacuolar proton-translocating ATPase. J Biol Chem 275(10): 6824–6830. 17. Poea-Guyon S, Amar M, Fossier P, Morel N (2006) Alternative splicing controls neuronal expression of v-ATPase subunit a1 and sorting to nerve terminals. J Biol Chem 281(25): 17164–17172. 18. Li YP, Chen W, Stashenko P (1996) Molecular cloning and characterization of a putative novel human osteoclast-specific 116-kDa vacuolar proton pump subunit. Biochem Biophys Res Commun 218(3): 813–821. 19. Heinemann T, Bulwin GC, Randall J, Schnieders B, Sandhoff K, et al. (1999) Genomic organization of the gene coding for TIRC7, a novel membrane protein essential for T cell activation. Genomics 57(3): 398–406. 20. Kawasaki-Nishi S, Yamaguchi A, Forgac M, Nishi T (2007) Tissue specific expression of the splice variants of the mouse vacuolar proton-translocating ATPase a4 subunit. Biochem Biophys Res Commun 364(4): 1032–1036. 21. Crider BP, Andersen P, White AE, Zhou Z, Li X, et al. (1997) Subunit G of the vacuolar proton pump. molecular characterization and functional expression. J Biol Chem 272(16): 10721–10728. 22. Neville MJ, Campbell RD (1999) A new member of the ig superfamily and a VATPase G subunit are among the predicted products of novel genes close to the TNF locus in the human MHC. J Immunol 162(8): 4745–4754. 23. Zhou Z, Peng SB, Crider BP, Slaughter C, Xie XS, et al. (1998) Molecular characterization of the 50- and 57-kDa subunits of the bovine vacuolar proton pump. J Biol Chem 273(10): 5878–5884. 24. Sayers EW, Barrett T, Benson DA, Bryant SH, Canese K, et al. (2009) Database resources of the national center for biotechnology information. Nucleic Acids Res 37(Database issue): D5–15. 25. Eyre TA (2006) The HUGO gene nomenclature database, 2006 updates. Nucleic Acids Res 34(90001): D319–D321. 26. Mcginnis S, Madden TL (2004) BLAST: At the core of a powerful and diverse set of sequence analysis tools. Nucleic Acids Res 32(Web Server): W20–W25. 27. Stajich JE (2002) The bioperl toolkit: Perl modules for the life sciences. Genome Res 12(10): 1611–1618. 28. Ryan MC, Zeeberg BR, Caplen NJ, Cleland JA, Kahn AB, et al. (2008) SpliceCenter: A suite of web-based bioinformatic applications for evaluating the impact of alternative splicing on RT-PCR, RNAi, microarray, and peptidebased studies. BMC Bioinformatics 9(1): 313. 29. Peng SB, Li X, Crider BP, Zhou Z, Andersen P, et al. (1999) Identification and reconstitution of an isoform of the 116-kDa subunit of the vacuolar proton translocating ATPase. J Biol Chem 274(4): 2549–2555. 30. Nishigori H, Yamada S, Tomura H, Fernald AA, Beau MML, et al. (1998) Identification and characterization of the gene encoding a second proteolipid subunit of human vacuolar H+-ATPase (ATP6F). Genomics 50(2): 222– 228. 31. Mandel M, Moriyama Y, Hulmes JD, Pan YC, Nelson H, et al. (1988) cDNA sequence encoding the 16-kDa proteolipid of chromaffin granules implies gene duplication in the evolution of H+-ATPases. Proc Natl Acad Sci USA 85(15): 5521–5524. 32. Wang SY, Moriyama Y, Mandel M, Hulmes JD, Pan YC, et al. (1988) Cloning of cDNA encoding a 32-kDa protein. an accessory polypeptide of the H+ATPase from chromaffin granules. J Biol Chem 263(33): 17638–17642. 33. Ludwig J, Kerscher S, Brandt U, Pfeiffer K, Getlawi F, et al. (1998) Identification and characterization of a novel 9.2-kDa membrane sectorassociated protein of vacuolar proton-ATPase from chromaffin granules. J Biol Chem 273(18): 10939–10947. 34. Wang ZQ, Gluck S (1990) Isolation and properties of bovine kidney brush border vacuolar H+-ATPase. A proton pump with enzymatic and structural differences from kidney microsomal H+-ATPase. J Biol Chem 265(35): 21957–21965. 35. Sudhof TC, Fried VA, Stone DK, Johnston PA, Xie XS (1989) Human endomembrane H+ pump strongly resembles the ATP-synthetase of archaebacteria. Proc Natl Acad Sci USA 86(16): 6067–6071. 36. Bernasconi P, Rausch T, Struve I, Morgan L, Taiz L (1990) An mRNA from human brain encodes an isoform of the B subunit of the vacuolar H+-ATPase. J Biol Chem 265(29): 17428–17431. 37. Nelson H, Mandiyan S, Noumi T, Moriyama Y, Miedel MC, et al. (1990) Molecular cloning of cDNA encoding the C subunit of H+-ATPase from bovine chromaffin granules. J Biol Chem 265(33): 20390–20393. 38. Kennell JA, Richards NW, Schaner PE, Gumucio DL (2001) cDNA cloning, chromosomal localization and evolutionary analysis of mouse vacuolar ATPase subunit D, Atp6m. Cytogenet Cell Genet 92(3–4): 337–341.

PLoS ONE | www.plosone.org

4

March 2010 | Volume 5 | Issue 3 | e9531

V-ATPase Subunit Genes

39. Hirsch S, Strauss A, Masood K, Lee S, Sukhatme V, et al. (1988) Isolation and sequence of a cDNA clone encoding the 31-kDa subunit of bovine kidney vacuolar H+-ATPase. Proc Natl Acad Sci USA 85(9): 3004–3008. 40. Imai-Senga Y, Sun-Wada GH, Wada Y, Futai M (2002) A human gene, ATP6E1, encoding a testis-specific isoform of H+-ATPase subunit E. Gene 289(1–2): 7–12. 41. Fujiwara T, Kawai A, Shimizu F, Hirano H, Okuno S, et al. (1995) Cloning, sequencing and expression of a novel cDNA encoding human vacuolar ATPase (14-kDa subunit). DNA Research : An International Journal for Rapid Publication of Reports on Genes and Genomes 2(3): 107–111.

42. Supekova L, Sbia M, Supek F, Ma Y, Nelson N (1996) A novel subunit of vacuolar H+-ATPase related to the b subunit of F-ATPases. J Exp Biol 199(Pt 5): 1147–1156. 43. Strausberg RL, Feingold EA, Grouse LH, Derge JG, Klausner RD, et al. (2002) Generation and initial analysis of more than 15,000 full-length human and mouse cDNA sequences. Proc Natl Acad Sci USA 99(26): 16899–16903. 44. Hu RM, Han ZG, Song HD, Peng YD, Huang QH, et al. (2000) Gene expression profiling in the human hypothalamus-pituitary-adrenal axis and fulllength cDNA cloning. Proc Natl Acad Sci USA 97(17): 9543–9548.

PLoS ONE | www.plosone.org

5

March 2010 | Volume 5 | Issue 3 | e9531