Author's personal copy - Boston University · Author's personal copy Developmental and Comparative Immunology 35 (2011) 273 284 ... Received in revised form 24 August 2010 Accepted

This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

http://www.elsevier.com/copyright

Author's personal copy

Developmental and Comparative Immunology 35 (2011) 273–284

Contents lists available at ScienceDirect

Developmental and Comparative Immunology

journa l homepage: www.e lsev ier .com/ locate /dc i

The two suborders of chiropterans have the canonical heavy-chainimmunoglobulin (Ig) gene repertoire of eutherian mammals

John E. Butlera,∗, Nancy Wertza, Yaofeng Zhaob, Shuyi Zhangc, Yonghua Baob,Sara Bratschd, Thomas H. Kunze, John O. Whitaker Jr. f, Tony Schountzg

a Department of Microbiology, University of Iowa, Iowa City, IA, United Statesb State Key Laboratory of Agrobiotechnology, China Agricultural University, Beijing, Chinac School of Life Sciences, East China Normal University, Shanghai, Chinad Department of Biology, University of Wisconsin-River Falls, United Statese Center for Ecology and Conservation Biology, Department of Biology, Boston University, Boston, MA, United Statesf Department of Biology, Indiana State University, Terra Haute, IN, United Statesg School of Biological Sciences, University of Northern Colorado, Greeley, CO, United States

a r t i c l e i n f o

Article history:Received 23 July 2010Received in revised form 24 August 2010Accepted 25 August 2010Available online 9 September 2010

Keywords:BatImmunoglobulin genesPhylogeny

a b s t r a c t

Bats comprise 20% of all mammals, yet little is known about their immune system and virtually nothingabout their immunoglobulin genes. We show that four different bat species transcribe genes encodingIgM, IgE, IgA and IgG subclasses, the latter which have diversified after speciation; the canonical patternfor eutherian mammals. IgD transcripts were only recovered from insectivorous bats and were comprisedof CH1, CH3 and two hinge exons; the second hinge exon was fused to CH3. IgA in all species resembleshuman IgA2 with the putative cysteine forming the bridge to the light chain found at position 77. Sequencecomparisons yielded no evidence for a diphyletic origin of the suborders. Bats show no close similarity toanother mammalian order; the strongest association was with carnivores. Data reveal that CH diversityand VDJ and CDR3 organization are similar to other eutherian mammals, although the expressed VH3family repertoire was unusually diverse.

© 2010 Elsevier Ltd. All rights reserved.

1. Introduction

Among living vertebrates, bats and birds are unique in their abil-ity to fly, and this common feature sets them apart ecologicallyand physiologically from other groups. Bats are in some ways thenocturnal equivalent of birds, having evolved and radiated into adiversity of forms to fill many of the same niches. Bats have success-fully colonized almost every continental region on earth (exceptAntarctica) as well as many oceanic islands and archipelagos (Kunz,1982; Kunz and Lumsden, 2003). Bats impact the environment inseveral ways. Insectivorous species suppress insect populations(Cleveland et al., 2006; Kalka et al., 2008; Williams-Guillén et al.,2008; Jones et al., 2009), and plant-visiting bats serve as pollina-tors and disperse seeds (Fujita and Tuttle, 1991; Shilton et al., 1999;Muscarella and Fleming, 2007; Fleming et al., 2009). Insectivorousspecies like the common little brown bat (Myotis lucifugus) mayconsume their entire body mass each night in insects (Kurta et al.,

∗ Corresponding author at: Department of Microbiology, University of Iowa, 3550BSB, 51 Newton Road, Iowa City, IA 52242, United States. Tel.: +1 319 335 7776;fax: +1 319 335 9006.

E-mail address: [email protected] (J.E. Butler).

1989). Thus, they are the most important predators of nocturnalinsects and consume 5–10 times more insects than swallows andflycatchers. So-called “megabats” or “flying foxes” of the suborderYinpteropodidae are important pollinators and seed dispersers, butsometimes can be pests when they damage fruit crops.

Bats comprise the second largest order of mammals (next torodents) in total number of species and may exceed all other groupsin overall abundance (Kunz, 1982; Wimsatt, 1970). Of the >5400recognized species of mammals, bats comprise 20% (Teeling et al.,2002; Simmons, 2005). Bats are believed to have evolved early andmorphologically have changed very little over the past 52 millionyears (Hill and Smith, 1984; Simmons et al., 2008). Bats are dividedinto 18 families that comprise two major suborders: Yinpteropo-didae and Yangochiroptera. The former suborder includes thefamily Pteropodidae which comprises the so-called “megabats”or “flying foxes” and the families Rhinolophidae, Hipposideridae,Megadermatidae, Rhinopomidae and Crasonycyeteridae while theYangochiroptera includes all other families including the insec-tivorous “microbats” (Teeling et al., 2002). Except for the genusRousettes, all of the families Pteropodidae are non-echolators com-pared to all other species of both suborders. The appearance ofcertain megabats (flying foxes) and neurological features of the eyeand prestrinate cortex lead Pettigrew to propose that the order was

0145-305X/$ – see front matter © 2010 Elsevier Ltd. All rights reserved.doi:10.1016/j.dci.2010.08.011


274 J.E. Butler et al. / Developmental and Comparative Immunology 35 (2011) 273–284

diphyletic, i.e. the Yinpteropodidae arose from a primate/lemursstock while all other bat families evolved from the Insectivora(Pettigrew, 1986). This was an intriguing hypothesis because itimplied that mammalian flight evolved twice. However, advancesin molecular analyses, including what we report here, have largelylaid to rest the diphyletic hypothesis for the Chiroptera (Murphyet al., 2007; Teeling et al., 2005; Adkins and Honeycutt, 1991;Mindell et al., 1991). Bats were originally considered members ofthe superorder Archonta which also included primates, tree shrewsand flying lemurs (Gregory, 1910; Simpson, 1945). Recent analy-ses places bats closer to ungulates, whales, horses and carnivores(Teeling et al., 2002, 2005) [a clade referred to as Laurasiatheria].

All bat species can be vectors for human viral diseases(Messenger et al., 2003; Calisher et al., 2006; Wibbelt et al., 2009).There are >85 viruses known from bats (Calisher et al., 2006) andseveral are associated with animal and human epidemics (Wong etal., 2007). These include Hendra and Nipah viruses (horse, humansand swine; Mackenzie and Field, 2004) SARS (rodents, cattle swine,dogs, human; Dobson, 2005; Li et al., 2005; Mueller et al., 2007;Shi and Hu, 2008) Ebola (human and other primates; Swanepoelet al., 1996; Leroy et al., 2005) and West Nile viruses (horses andhumans; Lau et al., 2005). Recently there is concern over their rolein zoonotic diseases (Calisher et al., 2006; Wibbelt et al., 2009; Lauet al., 2005; Halpin et al., 2008). This includes white nose syndrome(WNS) which has caused unprecedented mortality in little brownbat populations in the Northeastern US (Blehert et al., 2009; Fricket al., 2010).

However, little is known about the immune system of bats,especially their immunoglobulins (Igs) and the manner which theygenerate their antibody repertoire.

The paucity of information on the bat immune system raisedthe question of whether their immune system might show anydeviation from other mammals that would suggest an adaptionfor flight. Birds have an immune system disparate from mammals.They lack IgD and IgE and both chicken IgG (IgY) and IgA differ indomain structure from their mammalian counterparts (Ratcliffe,1991). Furthermore, birds lack lymph nodes, have determinant Bcell lymphopoiesis and use somatic gene conversion (SGC) to gen-erate their antibody repertoire (Reynaud et al., 1989; McCormacket al., 1991). Since the work of Teeling et al. (2002, 2005) leaveslittle doubt that bats are indeed part of the mammalian lineage,it is unlikely that their immune system would resemble that ofbirds. However, there is considerable variation among mammalsregarding their antibody repertoire (Butler, 2006), thus investigat-ing the Ig genes of a mammalian order not previously studied,might uncover variations in the canonical mammalian Ig patternthat could be valuable in future functional studies involving theimmune system.

In the present study, we demonstrate that the Igs of batspossess the canonical pattern characteristic of other eutherianmammals, including apparent diversification of IgG into subclassesafter speciation. Given the sequences for both Yinpteropodidae andYangochiroptera were available, including fruit-eating and insec-tivorous species, we found no evidence for a polyphyletic originfor bats. However, we did find immunological evidence for thedivergence of the two suborders. Our analysis revealed a surprisingand unexplained diversity among expressed VH3 genes of the littlebrown bat, M. lucifugus, which raises the question of whether batsmight use SCG similar to that of birds, extensive somatic hypermu-tation (SHM) or combinatorial diversity to develop their specificantibody repertoire. The latter question has been addressed in aseparate study that shows extensive germline diversity but a sur-prising lack of SHM (Bratsch et al., in press).

2. Materials and methods

2.1. Source of materials

Spleen and other lymphoid tissues were removed from euth-anized little brown bats (M. lucifugus) that had been collectedat a maternity roost in eastern Massachusetts. The same tissueswere also collected from the big brown bat (Eptiscus fuscus) fromwestern Indiana. Both species are members of the family Ves-pertilionidae, suborder Yangochiroptera. The same tissues werecollected from the Seba’s short-nosed fruit bat, Carollia perspicillata(family Phyllostomidae suborder Yangochiroptera) as by-productsof viral research at Northern Colorado University. CDNA from theshort-nosed fruit bat Cynopterus sphinx (family Pteropodidae), wasprepared at the East China Normal University and at the ChinaAgricultural University. Apart from the latter, all samples were col-lected and frozen in liquid nitrogen and shipped on dry ice to theUniversity of Iowa.

2.2. Preparation of RNA and cDNA

RNA was extracted from pulverized frozen tissue or from TriZol-stored material following the manufacturer’s protocol (Invitrogen,Carlsbad, CA) and as previously described (Sun et al., 1998). RNAconcentration was determined spectrophotometrically using aNanoDrop 1000. CDNA was prepared using Superscript II as pre-vious described (Butler et al., 2006).

2.3. Recovery of bat Ig transcripts by PCR

Transcripts encoding bat IgM, IgD, IgG, IgA and IgE were recov-ered by PCR using degenerate primer sets. Table 1 indicates the

Table 1Primers used for recovery of bat heavy-chain Ig transcripts.

Isotype Forwarda Reverse Product lengthb

IgMPCR1 VH FR 1 5′ agctggtggagtctggagga

IgM CH4 gtggactttcctctcggtPCR2 VH FR 1 3′ actctcctgtgcagcctctg 1500

IgGPCR1 VH FR 1 5′ agctggtggagtctggagga

IgG3′UTRc tttatttcatgctggdcccgggPCR2 VH FR 1 3′ actctcctgtgcagcctctg 1400

IgDPCR1 VH FR 1 5′ agctggtggagtctggagga

IgD CH3 tgcctgacctcgcatgtgtagPCR2 VH FR 1 3′ actctcctgtgcagcctctg 1450

IgAPCR1 VH FR 1 5′ agctggtggagtctggagga

IgA CH3 tgacgtgggtgggtttacccgPCR2 VH FR 1 3′ actctcctgtgcagcctctg 1300

IgEPCR1 VH FR 1 5′ agctggtggagtctggagga

IgE CH3 aaagaagttctggatcaggcaggtgPCR2 VH FR 1 3′ actctcctgtgcagcctctg 1500

a FR1 = framework 1 of VH3 genes. FR1 3′ is a nested primer that targets a sequence downstream from FR1 5′ .b Products of expected length were obtained, cloned and sequenced.c All genes encoding C� genes share a common UTR sequence.


J.E. Butler et al. / Developmental and Comparative Immunology 35 (2011) 273–284 275

primers used and the domain sequences targeted by the primersemployed in first (PCR1) and second (PCR2) round PCR amplifica-tion and the expected length of the final PCR products. Since theVH3 family is considered most ancestral (Schroeder et al., 1990)we used the FR1-5′ primer based on the VH3 genes of swine andother species that have VH3 genes in their genome (Butler, 2006;Butler et al., 2006). A hemi-nested PCR system was used in which

the second round used the internal forward FR1-3′ primer (Table 1),but the same reverse primers that annealed to sequences encodingthe CH3, CH4 or UTR regions of the various isotypes. PCR productlength was evaluated on agarose gels and those of expected lengthwere excised from the gels and cloned into pCR2TOPO according tothe manufacturer’s protocol (Invitrogen, Carlsbad, CA). The ligationmixture was plated on L–B plates. Blue/white selection was used

Fig. 1. The deduced amino acid sequences of the heavy chains of IgM, IgD, IgE and IgA from four bats fit the canonical pattern for the corresponding human Ig isotypes. Aminoacid positions of interest that are referred to in the text, are in boldface and are numbered according to the IMGT system (Lefranc and LeFranc, 2001). Cysteines involved in theintradomain S–S loop are in bold with arrows indicating the loop. The cysteines involved in the covalent bond to the light chain, are also in bold and indicated as L–H. PotentialN-linked glycosylation sites are in boldface. Position in each domain start with 1 as is done in the IMGT system. The functional hinge of IgA is part of the 5′ portion of the CH2 domain. Dots indicate that the residue is the same as for the reference human sequence. Human IgA2 was used for comparison since the extended hinge of human IgA1 isunique among mammals. Gaps are inserted into the alignment to obtain the best possible alignment with the human sequence that was used as a reference. Bat sequenceshave been deposited in GenBank and the accession numbers and those for the human sequences that were used for comparison are as follows: M. lucifugus = IgA, HM134924;IgD, HM134925, HM134926; IgE, HM134927; IgM, HM134936, HM134937; E. fuscus = IgA, HM134938; IgD, HM134939, HM134940; IgE, HM134941; IgM, HM134943; C.perspicillata = IgA, HM134944; IgE, HM134945; IgM, HM134948; C. sphinx = IgA, HM134948, HM134949; IgE, HM134950; IgM, HM134953.



Fig. 1. (Continued ).

to recover colonies that were subsequently prepared for sequenceanalysis. The samples were then processed using a Qiagen Quicklyses kit (Valencia, CA) and digested with EcoR1 for 2 h to releasethe inserted DNA after which the product was again analyzed on anagarose electrophoretic gel to confirm that products of the expectedlength were recovered.

2.4. Sequence analysis

Clones processed as described above were sent to the coresequencing facility of the University of Iowa where they were ana-lyzed using an Applied Biosystems Model 3730 (96 capillary) DNAsequencer. Our data are presented using the IMGT system in whichthe start codon for each domain is number 1 (Fig. 1). We have noevidence for variants at each position that are described for humanvariants on the IMGT website http://www.imgt.org.

2.5. Recovery of genomic IgD

Sequences from expressed IgD from M. lucifugus were usedto recover genomic IgD sequences (scafford 15451, 11.5 kb) byapplying BLAST to the partially sequenced genome of M. lucifu-gus (http://www.ensembl.org/Myotis lucifugus/Info/Index). A newprimer was made by aligning the sequences from the transcribedIgD from M. lucifugus with the sequence recovered from BLAST toidentify heavily conserved areas of bat IgD. This new primer recov-ered IgD from E. fuscus spleen cDNA.

2.6. Analysis of VDJ transcripts

VDJ sequences, generated as by-products of efforts to identifythe various Ig heavy-chain genes, were analyzed with regard to therecovery of identical VH genes (Table 2) and for the characteris-tics of their CDR3 regions (Table 3). Sixty-six IgG VDJ transcriptsand seven VDJ transcripts for IgM, IgD, IgE and IgA were recoveredfrom the spleen of M. lucifugus. In addition, 51 VDJ transcripts werealso recovered from E. fuscus, C. perspicillata and C. sphinx duringthe procedure to recover their genes transcribing IgM, IgD, IgG, IgAand IgE. These transcripts allowed for a partial characterization oftranscribed VDJs.

2.7. IgG subclass analysis

Putative IgG subclass/allotype sequences recovered from all fourbat species were compared in PileUp. The translated sequenceswere then grouped and aligned based on the PileUp so that majordifference within and among species could be identified. Becausemajor differences were identified in the hinge region, intraspe-cific comparisons of hinge sequences for different bat IgGs and forhuman, mouse, cattle and swine IgG subclasses were compared.

2.8. Phylogenetic comparisons of immunoglobulin heavy-chainconstant regions genes

Phylogenetic trees were constructed with PHYLIP 3.67 softwarecombined with the TreeView package (Felsenstein, 1989; Page,



Table 2Recovery of identical FR4 sequences was common while recovery of identical transcribed VH genes was not.

VH CDR1 CDR2 JH-FR4

Total Identical Identical Sequence Total Sequence Total Sequence

M. lucifugus 73 0

6 SYWMH 4 AISDNGVKT 29 WGQGTMVTVSS

5 SSWMNa 3 AINPNGGKI 20 WGQGTLVTVSSb

4 SYSMD 2 ADSGGST 7 WGPGTLVTVSS

3 DSWMN 2 LINVDGSTT 5 WGPGTLVTVSQ

3 TSWMN 2 AISTNGGNT 12 Unique

2 DYHMN 2 RINWNGGYI

2 NSWMN 2 YINSGSSYI

2 SYGMS 2 YISSGSSYI

2 SYPMN 54 Unique

2 SYWMD

2 SYWMS

40 Unique

E. fuscus 20 0

2 SHYMS 20 Unique 11 WGQGTMVTVSS

18 Unique 3 WGQGTQVTVSSc

2 WGQGTLVTVSSb

4 Unique

C. perspicillata 16 02 NYWMN 2 INTGGDSTY 12 WGQGTLVTVSSb

14 Unique 14 Unique 2 WGQGTQVTVSSc

2 Unique

C. sphinx 15 02 SSWMNa 15 Unique 13 WGQGTLVTVSSb

13 Unique 2 Unique

a Same CDR1 (underlined) found in two different species.b The same JH was found in all four species of bats (also underlined).c JH found in two of the bat species examined.

1996). Multiple sequence alignments for tree construction weregenerated using CLUSTAL. Prodist and Neighbor programs in thePHYLIP package were also used for tree construction. A consensustree was taken from 1000 bootstrapped phylogenetic trees.

3. Results

3.1. Bats possess the canonical isotype repertoire of eutherianmammals

Fig. 1 provides an alignment of the deduced amino acidsequences of IgM, IgD, IgE and IgA from four bat species with thatof the corresponding isotypes from human. Given that IgG isotypes

from bats appear to have diverged prior to the appearance of themajor mammalian orders (Fig. 2), their sequences are separatelycompared (see below). The major motifs for the five human isotypes(and other eutherian mammals; data not shown) are present in allfour bat species investigated in the present study. For those shownin Fig. 1, the intrachain disulfide loop in each domain is also presentin the corresponding domains of bat Igs and the sequences adjacentto the cysteines involved in these bridges are generally conserved.With some exceptions, the major potential N-glycosylation sites areconserved for each isotype in bats. In the case of IgM and IgA, theseare identical. This also appears to be the situation for IgD althoughthe CH2 domain of bat IgD is absent (Figs. 1 and 5). IgE is most diver-gent by this criterion since the N-glycosylation site in CH1 (position

Table 3CDR3 sequences motifs and CDR3 lengths in VDJ recovered with various heavy-chain transcripts.

Species Isotype Sequence of CDR3 Frequency CDR3 length (range) Mean CDR3 length

M. lucifugus

IgM Ra DRLGSFDV Wa 4 8–10 8.5IgA R YGAGADLAY W 5 6–12 9.2IgE R DRSTTVTDMEY W 3 6–12 10IgD R HGSYGDFDV W 2 9–10 9.5IgG R DWGHGRFDV W 66 4–12 8.6

E. fuscus

IgM R VYCASGICYGFDY W 3 13 10IgA R DGWGTYYLDY W 5 6–11 8.8IgE R DVRGIYDY W 3 7–8 7.7IgD R PPPLVAGGFDY W 4 11–13 11.5IgG R EGDAAGGFDY W 9 4–13 9.4

C. sphinx

IgM R EYQGFIGY W 1 8 8IgA R GGGAYNGMDY W 4 6–14 9.5IgE R DLTDLGY W 1 7 7IgDb

IgG R EGTTNTSPFGY W 10 8–12 9.9

C. perspicellata

IgM R DLNWDGY W 5 7–9 8IgA R GMGRYDFYY W 4 4–14 8.8IgE R GVVGGRYFDN W 2 10–12 11IgDb

IgG R SRYYSYIYSY W 7 4–12 9.6

a The boundaries of CDR3 between the terminal arganine (R) of the FR3 of VH3 family genes and the invariant tryptophan of FR4 (W) are indicated.b IgD has not been recovered from the cDNA of C. sphinx and C. perspicillata.



Fig. 2. Genes encoding the heavy-chain genes of IgM, IgD, IgE and IgA of bats comprise a separate group with highest similarity with carnivores. The phylogenetic tree wasconstructed by using amino acid sequences of the first and last CH domains (platypus � CH7 was used as the last domain for comparison) of these heavy-chain isotype.The heavy-chain constant regions are indicated with the appropriate Greek letters. M = M. lucifugus, E = E. fuscus, Ca = C. perspicillata; Cy = C. sphinz. For the convenience ofreaders, the regions dealing with these four isotypes are indicated in boxes. The values supporting each node are derived from 1000 bootstrapped phylogenetic trees. Theaccession numbers for the sequences used are for C�: cattle, AF109167; dog, L36871; dolphin, AY621035; echidna, AF416951; horse, AY247966; human, J00220; mouse,J00475; opossum, AF012110; panda, AY818387; swine, U12594; platypus, AY055778; rabbit, X51647; rat, AABR03049801; sheep, AF024645. C�: cattle, AF515672; dog,DQ297185; horse, AY631942; human, BC021276; mouse, V00786 and V00788; panda, AY818394; swine AF411239; platypus, EU503149; rat, AAO19643; sheep, AF411238.C: cattle U63640; dog, L36872; echidna, AY099258; horse, AJ305046; human, J00222; mouse, X01857; opossum, AF035194; panda, AY818389; swine, U96100; platypus,AY055780; rat, K02901; sheep, M84356. C�: cattle, U63637; camel, AB091831; cat, AB016712; dog, EF207721 and EF207724; dolphin, AAG40853; echidna, AF416952; horse,L49414; human, X14940; mouse, V00818; opossum, AF012109; panda, AY818392; swine, U50149; platypus, AY168639; rabbit, J00666; rat, AABR03049259; sheep, X59994.Accession numbers for bat species are the same as those given in Figs. 1 and 3.



Fig. 3. The IgG subclasses of bats diversified after speciation as in other eutherian mammals. The phylogenetic tree was constructed based on the amino acid sequences ofthe first CH domain of all IgG subclasses which is highly species-specific. The designation for bat species is the same as for Fig. 2, i.e. M, E, C� and C�. These designationsare shown in boldface. The accession numbers for the sequences compared are: cattle (X16701, M36946, U63638), camel (AJ421266, AJ131945), dog (AF354264, AF354265,AF354266, AF354267), horse (AJ302055, AJ302056, AJ312379, AJ302058, AJ312380, AJ312381), human (J00228, J00230, X03604, K01316), mouse (J00453, V00825, V00763,J00479, X00915), opossum (AF035195), swine (U03778, U03779, U03782, EU372658, EU372657, EU372655), possum (AF157619), rabbit (L29172), rat (AABR03048905,AABR03049560, AABR03048905, AABR03049912), sheep (X69797, X70983). GenBank numbers for bats C� sequences are: M. lucifugus (HM134929, HM134930, HM134931,HM134932, HM134933, HM134934, HM134935), E. fuscus (HM134942), C .perspicillata (HM134946), C. sphinx (HM134951, HM134952).



Fig. 4. The IgG subclasses of bats, like other eutherian mammals, differ most in their hinge regions and hinge motifs are not shared among species. The amino acid sequencesof the hinge regions of IgG from different bat species and for subclasses of human, mice, cattle and swine are presented. Sequence motifs shared by the hinge regions ofdifferent subclasses within a species are underlined. The −1, −2, etc. associated with subclasses in cattle and humans indicate the use of alternative hinge exons. Superscriptsfor swine C� genes denote allotypic variants. The accession numbers for the cg hinge sequences are the same as those listed in the legend of Fig. 3.

21) is not present in any bat species we examined. Moreover, thepotential N-glycosylation motif in the CH2 (position 43) is miss-ing in all bats and only one of the three potential N-glycosylationsites in the CH3 domain of human IgE (position 65) is found in batIgE. Because of this observation we examined the IgE sequences ofseveral other mammalian IgEs including the duckbill platypus andthe Virginia opossum. We found that the N-glycosylation motif atposition 21 in CH1 was common to human, mouse, cattle and swinebut was missing in all bats, in the opossum and the platypus. Cattlehave an additional site at position 22.

Of particular interest is the absence of a cysteine in the CH1domain of bat IgA that is part of the covalent S–S bond to the lightchains in human IgA1. Thus we used human IgA2 as our referencebecause it also lacks this cysteine. Furthermore human IgA2 lacksthe unique 13 amino acid hinge of human IgA1 that is not foundin any other mammal. This means that bat IgA resembles humanIgA2 and that of cattle and swine with a cysteine at position 77in CH 1 that is part of the putative L–H bridge, whereas IgA fromthe Virginia opossum and the duckbill platypus apparently use thecysteine at position 14 in CH1 to form the L–H bridge like in humanIgA1.

3.2. IGHC gene sequences of bats form a separate group mostsimilar to carnivores

Phylogenetic analysis of IgM, IgD, IgE and IgA gene sequencesfrom bats indicates that especially IgM and IgD share their great-est similarity to that of carnivores (Fig. 2). While IgE from bats alsoshares sequence homology with carnivores, it is also related to IgEfrom the horse, swine and the human; rodents and ruminant artio-dactyls are more distantly related. A somewhat similar pattern ofsequence similarity was found for bat IgA. Noteworthy is that thesequence similarity of IgE, IgM and IgA, which corresponds to theaccepted taxonomic relationship for the bats included in this study(Fig. 2). Given that IgM and IgD are the two most primordial Ig iso-types among vertebrates (Ohta and Flajnik, 2006) the comparison

of the C� and C� from bats to those in other mammals should bemost reliable. Using this criterion, a common ancestry of bats withcarnivores is suggested.

3.3. Subclass diversification occurred after speciation

The branch points for the various IgG subclasses appear to haveoccurred after species diversification (Fig. 3). Our data also showthat the IgG from the Neotropical, Seba’s short-tailed fruit bat, C.perspicillata and those from the Paleotropical, short-tailed fruit batC. sphinx form separate branches from the one leading to the insec-tivorous members of the Yangochiroptera, M. lucifugus and E. fuscus,of the family Vespertilionidae. On the basis of transcripts recovered,IgG in M. lucifugus has diversified to the greatest extent; seven dif-ferent transcripts were recovered, whereas the transcript of only asingle IgG was recovered from C. perspicillata. While some variantsmay be alleles, this cannot explain all the variants in M. lucifugus orC. sphinx because all sequences were recovered from tissues of thesame individuals.

3.4. IgG subclass genes in bats differ in the hinge region anddifferences appear order-specific

Alignment of the various C� transcripts recovered from dif-ferent bat species revealed that the putative bat IgG subclassgenes show their greatest difference in their hinge region. This fol-lows the pattern observed in other species (Butler et al., 2009).Thus, we focused our comparison on variations in the hinge ofthe various C� genes from the four species of bats available forthe present study. Since the nomenclature of subclasses in allspecies is based on order of discovery, same-name homology isnot implied. This should also be apparent from Fig. 3, which showsthat bat IgG subclasses diversified after speciation, as observed forother mammalian orders (Butler, 2006) making sequence compar-isons of subclasses between different species of little phylogeneticrelevance. Therefore, sequence similarities between the four bat



Fig. 5. The domain structure of IgD of bats is rodent-like but unique because of thefusion of the second hinge exon with the CH3 domain. The TM exons were missingin the retrieved genomic sequence. IgD was only recovered from two bat speciesand only the genomic domain structure of M. lucifugus is indicated.

species in the present study are postulated to reflect either conver-gent or parallel evolution.

The hinge sequences of IgG from different mammalian taxa,including bats, illustrate several features of evolutionary and func-tional importance. First, nearly all hinges are characterized by thepresence of a combination of prolines (often paired) in associationwith cysteines. Second, the actual motif for this feature appears tobe species- (or order-) specific; these are illustrated by the under-lined motifs in Fig. 4. In some cases different IgG subclasses sharenearly the same hinge. This is demonstrated in swine (Butler et al.,2009; Fig. 4) but has been observed in M. lucifugus, C. perspicillataand C. sphinx. The exception is in cattle.

Alternative and duplicated hinges also occur. This is best illus-trated by IgG3 in humans and cattle; this is also a common featureof the domain structure of IgD in some species. Since the genomicstructure of the bat C� genes has not yet been determined, this mayalso be true for some bat IgG subclasses.

3.5. The domain structure of bat IgD is unique but is otherwisemost similar to rodents

A partial genomic � sequence (scafford 15451, 11.5 kb) ofM. lucifugus was retrieved from its genome database using aBLAST-based approach. Although the typical transmembrane (TM)encoding exons are missing in the retrieved sequence, two con-stant region (C) domain encoding exons, could be identified withone corresponding to CH1, and the other to the CH3 of other knownplacental mammalian IgD heavy chains. The genomic structure ofIgD from a variety of vertebrates was compared with the partialgenomic structure of IgD from M. lucifugus (Fig. 5). These data showthat the domain structure of IgD from M. lucifugus is most similarto that of rodents in having only one hinge exon and in lacking aCH2 domain; H2 and CH3 are fused exons in M. lucifugus. Despitethis rodent-like feature, sequence similarities place bat IgD closerto that of carnivores than rodents (Fig. 2). In no eutherian mam-mals is there evidence for the multi-domain structure of IgD seen

in the duckbill platypus, the African clawed frog (Xenopus sp.) andthe catfish (Ohta and Flajnik, 2006; Zhao et al., 2006, 2009).

3.6. The transcribed VH3 repertoire is diverse

The primers used to recover the various C-region genes in batsinvolved the use of forward primers for the ancestral VH3 variablegene family. Thus, the 140 transcripts recovered all contain VDJsencoded by VH3 genes. Because VH3 family genes in all species dif-fer primarily in CDR1 and CDR2, our analysis focused on these genesegments. Sixty-one percent of the transcripts had CDR1 sequencesthat were only recovered once, 13% had CDR1s that were recoveredtwice and 4% had CDR1s that were recovered six times (Table 2).Being a longer sequence, a higher proportion of CDR2 sequenceswere unique (75%). By contrast only 14% of JH-FR4 region sequenceswere unique suggesting that there are a limited number of JH genesin the genome but many different VH3 genes.

Of the 73 VDJ sequences recovered form M. lucifugus, nonecontained identical VH genes when both CDR1 and CDR2 were con-sidered, although some of these VH genes share CDR regions withother VH genes. Sharing was most common among the shorter (15nucleotides) CDR1 segments. In one case an identical CDR1 fromM. lucifugus was also found twice in C. sphinx. By contrast, iden-tical FR4 regions (JH encoded) were quite frequently recovered.The number and sequences of the various JH segments is unknownsince the genome has not been mapped. In any case, the frequencyof FR4 recovery suggests that there are fewer JH than VH genesand that M. lucifugus probably has at least five JH gene segmentsin its genome. The limited diversity of JH is also supported by theobservation that all four bat species in the present study share oneidentical FR4. Since the species studied represent different fami-lies and suborders; (E. fuscus and M. lucifugus belong to the sameVespertilionidae, suborder Yangochiroptera, C. perspicillata to thefamily Phyllostomidae, suborder Yangochiroptera, and C. sphinx tothe family Pteropodidae, suborder Yinpterochiroptera) this JH geneis evolutionarily conserved.

All VDJ sequences were from transcripts so that the diversityamong these VH3 genes may be the result of somatic hypermuta-tion (SHM) of CDR1 and CDR2 regions. This issue has been recentlyaddressed in a separate study (Bratsch et al., in press).

3.7. The CDR3 regions of VDJ transcripts resembles that of mostmammals

The CDR3 portion of transcribed VDJs mostly results from con-tributions by DH segments, differences in the 5′ end of JH genesegments, nucleotide additions and SHM. Table 3 summarizes datafrom the most frequently encountered CDR3 regions transcribedwith different isotypes of the four bat species. Our data indicatethat the mean length of CDR3 was 9.1 ± 1.1 amino acids; the samelength found in mice, humans and swine. We found no significantdifferences in CDR3 length among the isotype or the various species.Identification of putative DH segments in CDR3 was not possiblebecause: (a) the same motif was never recovered twice among the140 transcripts and (b) there is currently no information availableon the genomic DH repertoire of bats.

4. Discussion

Biologists have long recognized the importance that 20% of theworld’s mammals play in the ecosystem. They suppress aerial insectpopulations, pollinate flowers and disperse seeds (Cleveland et al.,2006; Kalka et al., 2008; Williams-Guillén et al., 2008; Shilton etal., 1999; Fleming et al., 2009). Despite their ecological importance,interest in their immune system has been almost entirely limited tothe Old World family Pteropodidae (suborder Yinpterochiroptera).



Chakraborty and Chakravarty (1984) studied the immune system ofthe Indian flying fox (Pteropus giganteus) using a hemolytic plaqueassay and (Omatsu et al. (2003) purified IgG from the Egyptian fly-ing fox (Rousettus aegyptiacus, family Pteropodidae) and used it inserological comparison with other mammals. In studies on histo-plasmosis, McMurry et al. (1982) purified serum Igs from the NewWorld fruit bat, Artibeus lituratus and monitored changes duringinfection. However, to our knowledge, no attention has been givento the Igs of insectivorous bats in the suborder Yangochiroptera, andno effort has been made to clone and characterize the heavy-chainconstant region Ig genes of any bat species.

The studies herein were undertaken for three reasons. First, byaddressing the growing concern for emerging diseases carried bybats, we reasoned that an understanding of their immune systemmight provide valuable insight that could indirectly aid in miti-gating these diseases. Bats carry >85 viruses and are vectors for avariety of human pandemic disease like SARS, Hendra and Ebola(Calisher et al., 2006). WNS, a fungal disease of hibernating bats inthe eastern US (Blehert et al., 2009; Gargas et al., 2009), has causedone of the most precipitous declines ever reported for North Amer-ican wildlife, that is threatening regional extinction (Frick et al.,2010). Second, we wished to determine whether bats might showdifferences in their Ig repertoire or its formation from that seenin other mammals. Finally, we reasoned that by comparing the Iggenes of different bats species with those of other mammals, wecould improve our understanding of mammalian phylogeny in gen-eral and chiropteran phylogeny in particular. Various hypotheseshave been proposed linking bats with various taxa including pri-mates, rodents, insectivores, carnivores and ungulates (Teeling etal., 2002, 2005). A relatively recent hypothesis even proposed thatthe two suborders (i.e. so-called mega- and microbats), evolvedfrom two separate lineages (Pettigrew, 1986)

Our findings are unambiguous in showing that Ig genes of batsfit the canonical pattern of eutherian mammals, neither birds norproto- or protherian mammals. All four bat species examined inthe present study transcribed genes for IgM, IgG, IgA and IgE andboth insectivorous species of the suborder Yangochiroptera tran-scribed IgD (Figs. 1 and 2) with a domain structure characteristicof eutherian (e.g. placental) mammals but not of Protheria (e.g.monotremes) or lower vertebrates (e.g. amphibians and fish; Fig. 5).We were unable to recover a transcript for IgD from either C. per-spicillata or C. sphinx. IgD in mammals shows the least interspecifichomology among all Ig isotypes and has therefore been elusive torecover (Butler, 2006; Butler et al., 1996). It is likely that whengenomic projects for bats other than M. lucifugus appear, IgD mayalso be found in the genome of other bat species, with an alteredsequence and possibly as pseudogenes. Since IgD has been lost dur-ing evolution in some cases such as in birds, marsupials and eventhe eutherian rabbit (Mage et al., 2006), its loss in one suborder ofbats may not be surprising.

The differences among mammals in the existence of the cys-teine at position 15–18 (Fig. 1) in the CH1 domain that forms abridge to the light chain in human IgA, suggests that either bats,human IgA2, cattle and swine: (a) have light chains that are non-covalently bonded to the C� chain or (b) the cysteine at position 77(Fig. 1) is used in the L–H bridge. Because there is no biochemicalevidence of non-covalent L chains in swine and cattle, the latterhypothesis is most likely. Our results suggest that in the diversifi-cation of mammalian IgA, the pathways diverged in C�1 resultingin the human being the only mammal outside of the exceptionalrabbit family, which has more than one subclass and retains bothmotifs for covalent linkage to light chains.

Motifs for N-glycosylation are less likely to be conserved thancysteine resides but may nevertheless distinguish evolutionarypathways. In this respect, the pattern of N-glycosylation sites inall bats studied to date resembles that of the Virginia opossum

and duckbill platypus, but not other mammals for which data areavailable.

The diversification of IgG into subclasses is primarily a featureof eutherian mammals; the Virginia opossum has only a single IgGand the duckbill platypus has two very similar IgGs (Zhao et al.,2009; Wang et al., 2009). The IgG-like Igs in other vertebrates haveonly distant sequence similarities to those in eutherian mammals(Fig. 3). The IgG-like IgY of the chicken and lizard also shows nosubclass diversification (Wei et al., 2009), although in the Africanclawed frog (Xenopus sp.) a short Ig molecule, IgF, appears to havebeen derived from IgY and further diversified in genus Xenopus(Zhao et al., 2006). IgG has been regarded as the “flagship Ig” ofmammals and is not found in other vertebrates (Butler et al., 2009).Thus, its presence in bats and in diversified form (subclasses) fur-ther supports the conclusion that bats possess the canonical isotyperepertoire of other eutherian mammals.

Also consistent with the pattern seen in eutherian mammals isthat post-speciation diversification of IgG into subclasses mostlyaffects the sequence of the hinge exon (Butler et al., 2009). Weobserved the same to be true in the Chiroptera. The hinge sequencesof the IgG variants from bats, together with that of other eutherianmammals, are shown in Fig. 4. These data show that specific hingesequence motifs are often shared among the hinges of different IgGsin one species but not shared between species. Seven variants wererecovered from a single specimen of M. lucifugus, so that assum-ing this individual was heterozygous, there must be 3–4 subclasseugenes while the remainder could be alleles of these subclasses.At this time only a single IgG has been recovered from Seba’s short-tailed fruit bat (family Phyllostomidae, suborder Yangochiroptera;Fig. 3).

IgGs with extended hinges such as human and swine IgG3 arebest-suited for complement activation (Butler et al., 2009; Lefrancand LeFranc, 2001). In humans all but IgG2 are actively transportedacross the placenta (Lefranc and LeFranc, 2001) and IgG1 of cat-tle is selectively transported across the acinar epithelial cells ofthe mammary gland (Hammer et al., 1969; Brandon et al., 1971).Subclass IgGs also differ in their affinity for Fc�R1, Fc�RII andFc�RIII that are distributed among phagocytic cells and lympho-cytes, resulting in a further division of labor among subclasses(Ravetch and Kinet, 1991).

Differential subclass expression can depend on whether apro- or anti-inflammatory cytokine microenvironment is present(Mossman and Coffman, 1989). Based on the known divisionof function among IgG subclasses in well-studied species, theexpressed subclass diversity we report here for bats predicts a sim-ilar diversification of function for bat IgGs like that seen amonghuman IgG subclasses and in other eutherian mammals.

The organization of the heavy-chain variable region appearssimilar to that of other eutherian mammals. However, the low fre-quency of recovering a particular VH gene during a second attemptin >70 chances in M. lucifugus (Table 2) suggests an exceptionallydiverse germline VH3 repertoire or a high rate of somatic hypermu-tation. The same applies to the recovery of putative DH segments(Table 3). A separate study indicates that diversity is germlineencoded and not the result of extensive SHM (Bratsch et al., inpress).

As in laboratory mice, the rabbit family and humans, at least fivedifferent FR4 regions were recovered suggesting as many as five JHregions. Furthermore one of these is highly conserved since it wasfound in all bat species examined in this study and was recovered20 and 13 times respectively, in species from two different subor-ders (Table 2). CDR3 length was in the same range as in humans,laboratory mice and swine with no evidence for the unusually longCDR3 regions that are seen in cattle (Berens et al., 1997) or camels(Nguyen et al., 2000). Data summarized in Tables 2 and 3 thereforesuggests a VH, DH and JH repertoire at least similar to humans and



laboratory mice. These results are in stark contrast to swine (Sunet al., 1998; Butler et al., 2006).

Regarding the issue of phylogenetic origins, our data offer no evi-dence to support the diphyletic origin of bats (Pettigrew, 1986) andthus are consistent with the conclusions reached by others (Mindellet al., 1991; Nikaida et al., 2000; Teeling et al., 2005). Regardingtheir relationship to other mammalian orders, the sequence sim-ilarity of the genes encoding IgM and IgD may be most valuablebecause these are regarded as phylogenetically the two oldest Iggenes. Using this criterion, bats are closely related to carnivores(Fig. 2). While sequence similarity of carnivores and bat IgD isstrong, the domain structure is more rodent-like due to the lackof a CH2 domain (Fig. 5). Analogous with IgA, the hinge exon of batIgD is fused to the CH3 domain. In the case of IgD, this has not beenreported in other mammals. When genes for IgA and IgE are con-sidered, a relationship with carnivores is still present but there isalso a relationship with hoofed mammals (e.g. ungulates), humansand the dolphin (Fig. 2). Our data linking Chiroptera, Carnivora andPerissodactyla (horses) with the Laurasiatheria is consistent withstudies using retroposon insertion (Nishihara et al., 2006) and non-coding intron sequences (Matthee et al., 2007). The relationship tothe Perissodactyla was also reported by Murphy et al. (2007). Thus,from our analysis there is no evidence to support the hypothesisthat bats evolved from flying primates or rodents. Since IgG diver-sified following the appearance of bats (Fig. 3) sequence similarityamong eutherian mammals provides no insight into phylogeny butsuggest that IgG diversification is convergent.

Results from our study unambiguously support the hypothesisthat the Ig genes of both the Yinpteropodidae and the Yan-gochiroptera have the same canonical pattern of organizationas in other eutherian mammals, and thus are not in support ofa diphyletic origin for the Chiroptera. Their surprisingly diversevariable heavy-chain diversity appears to be germline encoded(Bratsch et al., in press). All other factors predict that the humoralimmune system of bats will function in much the same manneras in well-studied species such as laboratory rodents. This meansthat the humoral immune responses of bats to their pathogens,the viruses they transmit and vaccines that may be developed, arelikely to follow the same pattern as in well-studied, non-volanteutherian mammals.

Acknowledgement

Research supported in part by NIH Contract AI25489.

References

Adkins, R.M., Honeycutt, R.L., 1991. Molecular phylogeny of the superorder Archonta.Proc. Natl. Acad. Sci. 88, 10317–10321.

Berens, S.J., Wylie, D.E., Lopez, O.J., 1997. Use of a single VH family and long CDR3sin the variable region of cattle Ig heavy chains. Int. Immunol. 9, 189–199.

Blehert, D.S., Hicks, A.C., Behr, M., Meteyer, C.U., Berlowski-Zier, B.M., Buckles, E.L.,Coleman, J.T.H., Darling, S.R., Gargas, A., Niver, R., Okoniewski, J.C., Rudd, R.J.,Stone, W.B., 2009. Bat white-nose syndrome: an emerging fungal pathogen?Science 323, 227.

Brandon, M.R., Watson, D.L., Lascelle, A.K., 1971. The mechanism of transfer ofimmunoglobulins into mammary secretions of cows. Aust. J. Exp. Biol. Med. Sci.49, 613–623.

Bratsch, S., Wertz, N., Chaloner, K., Kunz, T.H., Butler, J.E., in press. The little brownbat displays a highly diverse VH, DH and JH repertoire but little evidence ofsomatic hypermutation. Dev. Comp. Immunol., doi:10.1016/j.dci.2010.06.004.

Butler, J.E., Sun, J., Navarro, P., 1996. The swine immunoglobulin heavy chain locushas a single JH and no identifiable IgD. Int. Immunol. 8, 1897–1904.

Butler, J.E., Weber, P., Wertz, N., 2006. Antibody repertoire development in fetal andneonatal pigs. XIII. “Hybrid VH genes” and the pre-immune repertoire revisited.J. Immunol. 177, 5459–5470.

Butler, J.E., Wertz, N., Deschacht, N., Kacskovics, I., 2009. Porcine IgG: structure,genetics and evolution. Immunogenetics 61, 209–230.

Butler, J.E., 2006. Preface: why I agreed to do this. In: Butler, J.E. (Guest Ed.), AntibodyRepertoire Development. Developmental and Comparative Immunology, vol. 30,special ed., pp. 1–17.

Calisher, C.H., Childs, J.E., Field, H.E., Homes, K.V., Schountz, T., 2006. Bats: importantreservoir host of emerging viruses. Clin. Microbiol. Rev. 19, 531–545.

Chakraborty, A.K., Chakravarty, A.K., 1984. Antibody-mediated response in the bat,Pteropus giganteus. Dev. Comp. Immunol. 8, 415–423.

Cleveland, C.J., Frank, J.D., Federico, P., Gomez, I., Hallam, T.G., Horn, J., Lopez, J.,McCracken, G.F., Medellin, R.A., Moreno-Valdez, A., Sansone, C., Westbrook, J.K.,Kunz, T.H., 2006. Economic value of the pest control service provided by Brazilianfree-tailed bat in south-central Texas. Front. Ecol. Environ. 4, 238–243.

Dobson, A.P., 2005. What links bats to emerging infectious diseases? Science 310,628–629.

Felsenstein, J., 1989. PHYLIP—phylogeny inference package (Version 3.2). Cladistics5, 164–166.

Fleming, T.H., Geiselman, C., Kress, W.J., 2009. The evolution of bat pollination: aphylogenetic perspective. Ann. Bot. 104, 1017–1043.

Frick, W.F., Pollock, J.F., Hicks, A., Langwig, K., Reynolds, D.S., Turner, G., Butchowski,C., Kunz, T.H., 2010. An emerging disease causes regional population collapse ofa common North American bat species. Science 328, 679–682.

Fujita, M.S., Tuttle, M.D., 1991. Flying foxes (Chiroptera: Pteropodidae): threatenedanimals of key ecological and economic importance. Conserv. Biol. 5, 455–463.

Gargas, A., Trest, M.T., Christensen, M., Volk, T.J., Blehert, D.S., 2009. Geomycesdestructans sp. associated with bat white-nose syndrome. Mycotaxon 108,147–154.

Gregory, W.K., 1910. The orders of mammals. Bull. Am. Museum Nat. Hist. 27, 1–524.Halpin, K., Hyatt, A.D., Plowright, R.K., Epstein, J.H., Daszak, P., Field, H., Wang,

L., Daniels, P.W., 2008. Henipavirus Ecology Research Group 2007 Emerg-ing viruses: coming in on a wrinkled wing and a prayer. Clin. Infect. Dis.,44711–44717.

Hammer, D.K., Kickhoefen, B., Henning, G., 1969. Molecular classes and propertiesof antibodies in cattle serum and colostrum synthesized during the primary andsecondary response to protein antigens. J. Biochem. 6, 443–454.

Hill, J.E., Smith, J.D., 1984. Bats: A Natural History. Univ. Texas Press, Austin.Jones, G., Jacobs, D.S., Kunz, T.H., Willig, M.R., Racey, P.A., 2009. Carpe noctem: the

importance of bats as bioindicators. Endangered Species Res. 8, 93–115.Kalka, M.B., Smith, A.R., Kalko, E.K.V., 2008. Bats limit arthropods and herbivors in a

tropical forest. Science 320, 71.Kunz, T.H. (Ed.), 1982. Ecology of Bats. Plenum Press, New York.Kunz, T.H., Lumsden, L.F., 2003. Ecology of cavity and foliage roosting bats. In: Kunz,

T.H., Fenton, M.B. (Eds.), Bat Ecology. Univ. Chicago Press, pp. 3–89.Kurta, A., Bell, G.P., Nagy, K.A., Kunz, T.H., 1989. Energetics of pregnancy and lactation

in free-ranging little brown bats (Myotis lucifugus). Physiol. Zool. 62, 804–818.Lau, S.K.P., Woo, P.C., Li, K.C., Huang, Y., Tsoi, H.W., Wong, B.H., Wong, S.S., Leung, S.Y.,

Chan, K.H., Yuen, K.Y., 2005. Severe acute respiratory syndrome coronavirus-likevirus in Chinese horseshoe bats. Proc. Natl. Acad. Sci. U.S.A. 102, 14040–14045.

Lefranc, M.-P., LeFranc, G., 2001. The Immunoglobulin Facts Book. Academic Press,NY, pp. 457.

Leroy, E.M., Kumulungui, B., Pourrut, X., Rouquet, P., Hassmanin, A., Yaba, P., Delicat,A., Paweska, J.T., Gonzales, J.P., Swanepoel, P., 2005. Fruit bats as reservoirs ofEbola virus. Nature 438, 575–576.

Li, W., Shi, Z., Yu, M., Ren, W., Smith, C., Epstein, J.H., Wang, H., Eaton, G.B.T., Zhang, S.,Wang, L.F., 2005. Bats are natural reservoirs of SARS-like coronaviruses. Science310, 676–679.

Mackenzie, J.S., Field, H., 2004. Emerging encephalitogenic viruses: lyssaviruses andhenipaviruses transmitted by frugivorous bats. Arch. Virol. Suppl. 18, 97–111.

Mage, R.G., Lanning, D.D., Knight, K.L., 2006. B cell and antibody repertoire devel-opment in rabbits: the requirement of gut-associated lymphoid tissues. Dev.Comp. Immunol. 30, 137–153.

Matthee, C.A., Eick, G., Willows-Munro, S., Montgelard, C., Pardini, A.T., Robinson,T.J., 2007. Indel evolution of mammalian introns and the utility of non-codingnuclear markers in eutherian phylogenetics. Mol. Phylogenet. Evol. 42, 827–837.

McCormack, W.T., Tjoelker, L.W., Thompson, C.B., 1991. Avian B cell developmentalgeneration of an immunoglobulin repertoire by gene conversion. Annu. Rev.Immunol. 9, 219–241.

McMurry, D.N., Stroud, J., Murphy, J.J.R., Carlomagno, M.A., Greer, D.L., 1982. Roleof immunoglobulin classes in experimental histoplasmosis in bats. Dev. Comp.Immunol. 6, 557–567.

Messenger, S.L., Rupprecht, C.E., Smith, J.S., 2003. Bats, emerging virus infections,and the rabies paradigm. In: Kunz, T.H., Fenton, M.B. (Eds.), Bat Ecology. Univ.Chicago Press, Chicago, pp. 622–679.

Mindell, D.P., Dick, C.W., Baker, R.J., 1991. Phylogenetic relationships among mega-bats, microbats and primates. Proc. Natl. Acad. Sci. U.S.A. 88, 10322–10326.

Mossman, T.R., Coffman, R.L., 1989. THI and TH2 cells: different patterns of lym-phokines secretion lead to different functional properties. Annu. Rev. Immunol.7, 145–173.

Mueller, M.A., Paweska, J.T., Leman, P.A., Drosten, C., Grywna, K., Kemp, A., Raach,L., Sonnenberg, K., Niedrig, M., Swanepoel, R., 2007. Coronavirus antibodies inAfrican bat species. Emerg. Infect. Dis. 13, 1367–1370.

Murphy, W.J., Pringle, T.H., Crider, T.A., Springer, M.S., Miller, W., 2007. Usinggenomic data to unravel the root of the placental mammal phylogeny. GenomeRes. 174, 413–421.

Muscarella, R., Fleming, T.H., 2007. The role of frugivorous bats in tropical forestsuccession. Biol. Rev. Camb. Philos. Soc. 82, 573–590.

Nguyen, V.K., Hamers, R., Wyns, L., Muyldermanns, S., 2000. Camel heavy chainantibodies: diverse germline VHH and specific mechanisms enlarge the antigen-binding repertoire. EMBO J. 19, 921–931.

Nikaida, M., Harada, M., Cao, Y., Hasegawa, M., Okada, N., 2000. Monophyletic ori-gin of the order Chiroptera and its phylogenetic position among mammals, as



inferred from the composite sequences of the mitochondrial DNA of a Japanesemegabat, the Ryukyu flying fox (Pteropus dasymailus). J. Mol. Evol. 51, 318–328.

Nishihara, H., Hasegawa, M., Okada, N., 2006. Pegasoferae, an unexpected mam-malian clade revealed by tracking ancient retroposan insertions. Proc. Natl. Acad.Sci. U.S.A. 103, 9929–9934.

Ohta, Y., Flajnik, M., 2006. IgD like IgM is a primordial immunoglobulin class perpet-uated in most jawed vertebrates. Proc. Natl. Acad. Sci. U.S.A. 103, 10723–19728.

Omatsu, T., Ishii, Y., Kyuwa, S., Milanda, E.G., Terao, K., Yoshikawa, Y., 2003. Molecularevolution inferred from immunological cross-reactivity of immunoglobulin Gamong Chiroptera and closely-related species. Exp. Anim. 52, 425–428.

Page, R.D., 1996. TreeView: an application to display phylogenetic trees on personalcomputers. Comput. Appl. Biosci. 12, 357–358.

Pettigrew, J.D., 1986. Flying primates? Megabats have advanced pathway from eyeto midbrain. Science 231, 1304–1306.

Ratcliffe, M.J.H., 2006. Antibodies, immunoglobulin genes and the bursa of Fabriciusin chicken B cell development. Dev. Comp. Immunol. 30, 101–118.

Ravetch, J.V., Kinet, J.P., 1991. Fc receptors. Annu. Rev. Immunol. 9, 457–492.Reynaud, C.-A., Dahan, A., Anquez, V., Weill, J.C., 1989. Somatic hyperconversion

diversifies the single VH gene of the chicken with a high incidence in the Dregion. Cell 59, 171–183.

Schroeder Jr., H.W., Hillson, J.L., Perlmutter, R.M., 1990. Structure and evolution ofmammalian VH families. Int. Immunol. 2, 41–50.

Shi, Z., Hu, Z., 2008. A review of studies on animal reservoirs of the SARS coronavirus.Virus Res. 133, 74–87.

Shilton, L.A., Altringham, J.D., Compton, S.G., Whittaker, R.J., 1999. Old World fruitbats can be long-distance seed dispersers through extended retention of viableseeds in gut. Proc. R. Soc. Ser. B 266, 219–223.

Simmons, N.B., 2005. Order Chiroptera. In: Wilson, D.E., Reeder, D.M. (Eds.), Mam-mal Species of the World: A Taxonomic and Geographic Reference, 2nd ed.Smithsonian Institution Press, Washington, DC, pp. 159–174.

Simmons, N.B., Seymour, K.L., Habersetzer, J., Gunnell, G.F., 2008. Primitive EarlyEocene bat from Wyoming and the evolution of flight and echolocation. Nature451, 818–821.

Simpson, G.G., 1945. The principles of classification and a classification of mammals.Bull. Am. Museum Nat. Hist. 85, 1–350.

Sun, J., Hayward, C., Shinde, R., Christenson, R., Ford, S.P., Butler, J.E., 1998. Anti-body repertoire development in fetal and neonatal piglets. I. Four VH genesaccount for 80% of VH usage during 84 days of fetal life. J. Immunol. 161, 5070–5078.

Swanepoel, R., Leman, P.A., Bart, F.J., Zachariades, N.A., Braack, L.E., Ksiazek, T.G.,Rollin, P.E., Zaki, S.R., Peters, C.J., 1996. Experimental inoculation of plants andanimals with bola virus. Emerg. Infect. Dis. 2, 321–325.

Teeling, E.C., Madsen, O., Van Den Busche, R.A., deJong, W.W., Stanhope, M.J.,Springer, M.S., 2002. Microbat paraphyly and the convergent evolution of a keyinnovation in Old World rhinolophoid microbats. Proc. Natl. Acad. Sci. U.S.A. 99,1431–1436.

Teeling, E.C., Springer, M.S., Madsen, O., Bates, P., O’Brien, J.J., Murphy, W.J., 2005.A molecular phylogeny for bats illuminates biogeography and the fossil record.Science 307, 580–584.

Wang, X., Olp, J.J., Miller, R.D., 2009. On the genomics of immunoglobulins in the gray,short-tailed opossum Monodelphis domestica. Immunogenetics 61, 581–596.

Wei, Z., Wu, Q., Ren, L., Hu, X., Guo, Y., Warr, G.W., Hammarstrom, L., Li, N., Zhao, Y.,2009. Expression of IgM, IgD and IgY in a reptile, Anolis carolinensis. J Immunol183, 3858–3864.

Wibbelt, G., Speck, S., Field, H., 2009. Methods for assessing diseases in bats. In: Kunz,T.H., Parsons, S. (Eds.), Ecological and Behavioral Methods for the Study of Bats.Johns Hopkins University Press, Baltimore, pp. 775–793.

Williams-Guillén, K., Perfecto, I., Vandermeer, J., 2008. Bats limit insects in aNeotropical agroforestry ecosystem. Science 320, 703.

Wimsatt, W.A., 1970. Biology of Bats, vol. 1. Academic Press, NY.Wong, S., Lau, S., Woo, P., Yuen, K.Y., 2007. Bats as a continuing source of emerging

infections in humans. Res. Med. Virol. 17, 67–91.Zhao, Y., Ciu, H., Whittington, C.M., Wei, Z., Zhang, X., Zhang, Z., Yu, L., Ren, L., Hu, X.,

Zhang, Y., Hellman, L., Belov, K., Li, N., Hammarstrom, L., 2009. Ornithorhynchusanatinus (Platypus) links the evolution of immunoglobulin genes in eutherianmammals and nonmammalian tetrapods. J. Immunol. 183, 3285–3293.

Zhao, Y., Pan-Hammarstrom, Q., Yu, S., Wertz, N., Zhang, X., Li, N., Butler, J.E., Ham-marstrom, L., 2006. Identification of IgF, a hinge-region containing Ig class andIgD in Xenopus tropicalis. Proc. Natl. Acad. Sci. U.S.A. 103, 12087–12092.

Documents

Author's personal copy - Boston University · Author's personal copy Developmental and Comparative Immunology 35 (2011) 273 284 ... Received in revised form 24 August 2010 Accepted