Nanobodies: Natural Single-Domain Antibodies€¦ · Nanobodies: Natural Single-Domain Antibodies Serge Muyldermans Research Group, Cellular and Molecular Immunology, Vrije Universiteit

BI82CH17-Muyldermans ARI 4 March 2013 19:59

RE V I E W

S

IN

AD V A

NC

E

Nanobodies: NaturalSingle-Domain AntibodiesSerge MuyldermansResearch Group, Cellular and Molecular Immunology, Vrije Universiteit Brussel,1050 Brussels, Belgium; email: [email protected]

Department of Structural Biology, VIB, Vrije Universiteit Brussel, 1050 Brussels, Belgium

Annu. Rev. Biochem. 2013. 82:17.1–17.23

The Annual Review of Biochemistry is online atbiochem.annualreviews.org

This article’s doi:10.1146/annurev-biochem-063011-092449

Copyright c© 2013 by Annual Reviews.All rights reserved

Keywords

camel, llama, shark, heavy-chain antibody, VHH, single-domainantibody

Abstract

Sera of camelids contain both conventional heterotetrameric antibod-ies and unique functional heavy (H)-chain antibodies (HCAbs). The Hchain of these homodimeric antibodies consists of one antigen-bindingdomain, the VHH, and two constant domains. HCAbs fail to incorpo-rate light (L) chains owing to the deletion of the first constant domainand a reshaped surface at the VHH side, which normally associates withL chains in conventional antibodies. The genetic elements composingHCAbs have been identified, but the in vivo generation of these antibod-ies from their dedicated genes into antigen-specific and affinity-maturedbona fide antibodies remains largely underinvestigated. However, thefacile identification of antigen-specific VHHs and their beneficial bio-chemical and economic properties (size, affinity, specificity, stability,production cost) supported by multiple crystal structures have encour-aged antibody engineering of these single-domain antibodies for use asa research tool and in biotechnology and medicine.

17.1

Review in Advance first posted online on March 13, 2013. (Changes may still occur before final publication online and in print.)

Changes may still occur before final publication online and in print

Ann

u. R

ev. B

ioch

em. 2

013.

82. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by Y

ale

Uni

vers

ity -

SO

CIA

L S

CIE

NC

E L

IBR

AR

Y o

n 03

/19/

13. F

or p

erso

nal u

se o

nly.


Immunoglobulin-γ(IgG): the mostabundant antibodycirculating in blood ofmammals

Contents

INTRODUCTION . . . . . . . . . . . . . . . . . . 17.2STRUCTURAL FEATURES OF

HEAVY-CHAIN ANTIBODIESAND NANOBODIES . . . . . . . . . . . . . 17.3Heavy-Chain Antibody Subisotypes

in Sera of Camelids . . . . . . . . . . . . . 17.3Structure of VHH . . . . . . . . . . . . . . . . . 17.4

DEDICATED HEAVY-CHAINANTIBODY GENES . . . . . . . . . . . . . 17.7Organization of the H Locus in the

Genome of Camelids . . . . . . . . . . . . 17.7Dedicated VHH Genes in the H

Locus . . . . . . . . . . . . . . . . . . . . . . . . . . 17.8Promiscuous VH Genes Contribute

to the VHH Repertoire . . . . . . . . . 17.9Heavy-Chain Antibody Ontogeny . .17.10

BIOCHEMICAL PROPERTIES OFNANOBODIES . . . . . . . . . . . . . . . . . . .17.10Identification of Antigen-Specific

Nanobodies . . . . . . . . . . . . . . . . . . . .17.10Affinity Parameters of Nanobodies . .17.11Expression of Recombinant

Nanobodies . . . . . . . . . . . . . . . . . . . .17.11Stability of Nanobodies . . . . . . . . . . . .17.11Nanobodies Are Nonimmunogenic .17.12Recognition of Unique Epiopes by

Nanobodies and Multivalentand/or Multispecific Constructs .17.12

APPLICATIONS WITHNANOBODIES . . . . . . . . . . . . . . . . . . .17.13Nanobodies as Research Tools . . . . .17.13Nanobodies as Diagnostic Tools . . . .17.14Nanobodies as Therapeutics . . . . . . . .17.14

INTRODUCTION

The overall structure of immunoglobulin-γ(IgG) antibodies assembled from two identi-cal heavy (H)-chain and two identical light(L)-chain polypeptides is well established andhighly conserved in mammals (Figure 1) (1).The L chain of these immunoglobulins com-prises two domains, whereas the H chainfolds into four domains. The sequence of the

N-terminal domain of the H and L polypep-tides’ chains varies between antibodies (desig-nated as variable domains, i.e., VH and VL).The paired VH-VL domains constitute thevariable fragment (Fv) that recognizes the anti-gen. The remaining H and L sequences aremore conserved (abbreviated as CH and CL,respectively). The two last CH regions are im-portant for recruitment of immune cells (e.g.,macrophages and natural killer cells) or for ef-fector functions (e.g., complement activation).

One notorious exception to this conven-tional mammalian IgG structure is found in seraof Camelidae (2). In addition to the conven-tional heterotetrameric antibodies, these serapossess special IgG antibodies. IgG antibod-ies, known as heavy-chain antibodies (HCAbs),are devoid of the L chain polypeptide and areunique because they lack the first constant do-main (CH1) (Figure 1). At its N-terminal re-gion, the H chain of the homodimeric proteincontains a dedicated variable domain, referredto as VHH, which serves to associate with itscognate antigen. The VHH in an HCAb is thestructural and functional equivalent of the Fabfragment (antigen-binding fragment) of con-ventional antibodies (Figure 1).

The biological family Camelidae comprisescamels (one-humped Camelus dromedariusand two-humped Camelus bactrianus), llama’s(Lama glama and Lama guanicoe), and vicugna’s(Vicugna vicugna and Vicugna pacos). Camelidaeis the only extant family in the suborderTylopoda, which constitutes together withRuminantia (bovine, goat, sheep, antelope, andothers) and Suiformes (pig and hippopotami)the order of Artiodactyla. Although all camelidshave HCAbs in their sera, other artiodactylssuch as Suiformes and Ruminantia do not havesuch functional HCAbs. However, HCAbsresulting from a genetic deletion of significantparts of the VH and CH1 regions have beenreported to occur in sera of humans witha pathological disorder (3, 4) or in mousehybridoma (5). Because of the truncated VHregion and absence of VL, these HCAbs arenonfunctional in antigen binding. Remarkably,immunoglobulins lacking L chains and devoid

17.2 Muyldermans


Ann

u. R

ev. B

ioch

em. 2

013.

82. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by Y

ale

Uni

vers

ity -

SO

CIA

L S

CIE

NC

E L

IBR

AR

Y o

n 03

/19/

13. F

or p

erso

nal u

se o

nly.


Heavy/light (H/L)chains: polypeptidechains of antibodieswith H and L chains,which have a mol wt of55,000 and 25,000,respectively

VH: the variabledomain of the heavychain ofimmunoglobulins

VL: the variabledomain of the lightchain ofimmunoglobulins

Heavy-chainantibody (HCAb): anantibody without lightchains; in the case ofcamelid IgG, the firstconstant domain isabsent in the H chains

of a conventional CH1 also occur in nurseshark, wobbegong, and perhaps spotted ratfish(6). These Ig-NAR ancestral antibodies have avariable domain, known as V-NAR, for antigenrecognition. Although the variable (V) se-quences of Ig-NARs and those of camel HCAbsare quite diverse, they show a surprising struc-tural and functional convergent evolution (7).

The antigen-binding fragment of an HCAb,containing one single variable domain, al-lows straightforward identification of antigen-binding VHHs after immunizing a camelid,cloning the VHH repertoire of B cells circu-lating in blood, and panning by phage display(8). The recombinant antigen-specific, single-domain VHH with dimensions in the nanome-ter range is also known as a nanobody (Nb) orsingle-domain antibody (sdAb). In this review,we summarize the main structural characteris-tics of HCAbs and the VHH domain, as wellas provide an update of the possible genera-tion mechanisms of HCAbs in Camelidae. Fi-nally, we review the biochemical properties ofthe Nbs that form the basis for a wide variety

−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−→Figure 1Schematic representation of naturally occurringantibodies in sera of camelids: conventionalantibodies (IgG1) containing two light (L) chains(the VL and CL domains) and two heavy (H) chains(composed of the VH, CH1, hinge, and CH2 andCH3 domains) and two types of homodimericheavy-chain antibodies (HCAbs), IgG2 and IgG3,which comprise only H chains; each H chaincontains a VHH, hinge, and CH2 and CH3domains. The hinge of the IgG2 fraction is longerthan that of the IgG3 type. As indicated in thefigure, the CH2 and CH3 domains form the Fc part.In IgG1, the first two domains of the H chain andthe L chain form the Fab fragment. The smallestintact functional antigen-binding fragment that canbe generated from conventional antibodies,consisting of a VH-VL pair and linked by anoligopeptide, is known as scFv (top right). Thesmallest intact functional antigen-binding fragmentof HCAbs is the single-domain VHH, also known asa nanobody (Nb) (bottom right). Abbreviations: CH,constant domain of immunoglobulin H chain; Fab,antigen-binding fragment; Fc, crystallizablefragment; scFv, single-chain variable fragment.

and an increasing number of routine and inno-vative applications in research, biotechnology,and medicine.

STRUCTURAL FEATURES OFHEAVY-CHAIN ANTIBODIESAND NANOBODIES

Heavy-Chain Antibody Subisotypesin Sera of Camelids

The absence of L chains in the IgG and lackof the CH1 in the H chain are key character-istics of camelid HCAbs. Consequently, theirsmaller size with a mol wt of 90,000 rather than150,000 for conventional antibodies and theirmore compact architecture might be betteradapted to access hidden targets. Conversely,

IgG2

VHH

VHH

VHH

VHH

Fc

IgG3

Fc

IgG1CH3 CH2

CH3 CH2

CH3 CH2

CH3 CH2

CH3 CH2

CH3 CH2

CL

VL

CH1

VH

FcCL

VH

CH1

VHH

VHVL

VL

scFv

Nb

www.annualreviews.org • Heavy-Chain Antibodies And Nanobodies 17.3


Ann

u. R

ev. B

ioch

em. 2

013.

82. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by Y

ale

Uni

vers

ity -

SO

CIA

L S

CIE

NC

E L

IBR

AR

Y o

n 03

/19/

13. F

or p

erso

nal u

se o

nly.


CH1: the firstconstant domain in theheavy chain ofantibodies

VHH:antigen-bindingvariable domain of theH chain ofheavy-chain antibodies

Fab: anantigen-bindingfragment ofconventionalantibodies, comprisingthe light chain, theVH, and the firstconstant domain of theheavy chain

Immunoglobulinnew antigen receptor(Ig-NAR):a homodimeric sharkantibody devoid oflight chains

Nanobody (Nb):the recombinantsingle-domain,antigen-specific VHHderived from a camelidHCAb

Single-domainantibody (sdAb):a fragment that can bederived from V-NAR,VHH, or human VH

Isotype: H or L chainantibody variants; forthe H chain, these areexpressed fromdedicated genes suchas IGHM or IGHGconstant genes

Complementarity-determining region(CDR): a loop invariable regions ofantibodies, encoded bythe IGHV gene orgenerated by V-D-Jgene recombination

the shorter distance between the two paratopeswithin one HCAb might compromise theircapacity to cross-link antigens, although ithas been speculated that the long IgG2 hingecontaining Pro-Gln repeats might form anextended spacer that structurally replaces theCH1 region (Figure 1) (2).

The percentage of HCAbs and conventionalIgG in the sera of camelids is variable: Incamels, it might reach 50–80%, whereas inSouth American camelid species, it totals up to10–25% (9). Such a significant proportion is atestament for the importance of HCAbs in theimmune protection of the camelid.

Several IgG subisotypes devoid of L chainscirculate in the bloodstream of camelids, al-though the exact number remains controversialowing to the difficulties of purifying subiso-types to homogeneity and of distinguishingsubisotypes from allotypes. Nevertheless, thehomodimeric and heterotetrameric antibodiesare easily separated by differential affinitychromatography on protein G and proteinA, where conventional antibodies are elutedat a lower pH from protein G columns(2, 10). The dromedary IgG fraction that is notretained on protein G is adsorbed on proteinA, and careful pH-controlled elution yieldstwo distinct HCAb-IgG2 fractions, namedIgG2a and IgG2b (10). The investigators showthat fractions from these adsorbents mightstill contain several subisotypes. Monoclonalantibodies raised against the IgGs of differentprotein A/G fractions and used to screen llamasera immunoglobulins confirmed the presenceof two isotypes of conventional IgGs as well aspossibly three and two subisotypes within theIgG2 and IgG3 fractions, respectively (11–13).

The analysis of IgG cDNA from dromedaryB lymphocytes reveals the occurrence of twoIgG subisotypes for conventional antibodies,whereas HCAbs include multiple IgG subiso-types (14). The various cDNA sequences stillneed to be matched to the subisotypes (or al-lotypes) as obtained from protein A and pro-tein G chromatography or anti-isotype mon-oclonal antibodies. Despite the uncertainty ofthe exact number of IgG subisotypes circulat-

ing in camelid blood, the numbers are withinthe range of IgGs in other artiodactyls. ThreeIgG isotypes have been identified from bovine(15), whereas pigs contain six IgG isotypes thatcannot be separated by protein A and/or proteinG chromatography (16).

Studies of human and mouse antibodiesdemonstrate a clear division of labor amongIgG subisotypes. Although knowledge of theexact roles and functions of the various camelidIgG subisotypes is still in its infancy, an infectedor vaccinated animal raises an immune responsein the three isotype fractions (to various extentsin different animals, depending on the actualimmunogen). Remarkably, the llama IgG1 andIgG3 neutralize West Nile virus, whereas IgG2seems less effective in this respect (17). How-ever, the same study indicated that IgGs fromall three isotypes bind to the surface of mono-cytes and macrophages, indicating that they areable to recruit immune cells.

Structure of VHH

The sequence variability within V domains islocalized in three hypervariable (HV) regionssurrounded by more conserved framework (FR)regions (Figure 2) (1). The folded V domaincomprises nine β-strands (A-B-C-C′-C′′-D-E-F-G), organized in a four-stranded β-sheet anda five-stranded β-sheet, connected by loops andby a conserved disulfide bond between Cys23and Cys94, packed against a conserved Trp. Inthis architecture, the HV regions are located inthe loops H1 to H3 that connect the B-C, theC′-C′′, and the F-G strands, respectively, andthat cluster at the N-terminal end of the domainforming a continuous surface, which is com-plementary to the surface of the epitope, henceits name complementarity-determining region(CDR). Although the sequence within the loopis HV, the length variation is limited exceptfor the H3 loop. Surprisingly the Cα positionswithin the H1 and H2 loops occupy restrictivelocations. Apparently only a few, so-called,canonical loop structures have been observedin human or mouse VH structures, and eventhe number of loop architecture combinations

17.4 Muyldermans


Ann

u. R

ev. B

ioch

em. 2

013.

82. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by Y

ale

Uni

vers

ity -

SO

CIA

L S

CIE

NC

E L

IBR

AR

Y o

n 03

/19/

13. F

or p

erso

nal u

se o

nly.


Gly

D

Arg

Phe

Glu

A B E

F C C’ C” G

IGHMJDIGHV3 IGHAIGHEn n n n

IGHV4IGHV1

VH V42

G49

L5

0 W

52

CDR1 CDR2 CDR3

FR1 FR2 FR3 FR4 F4

2 E4

9 R5

0 G

52

CDR1

CDR2

CDR3

VHH

IGHG1IGHG2

Genome H-locus

n n

CH1 CH2 CH3

CH1 CH2 CH3

W11

8

IGHV3H

Trp

Figure 2Schematic representation of the H locus in the genome of camelids consists of several variants of the IGHV1, IGHV4, IGHV3, andIGHV3H genes upstream; there is a cluster of D genes and a cluster of J gene segments, followed by the constant genes IGHM andIGHD (not shown), several IGHG subisotypes, and IGHE and IGHA genes. During B-cell lymphopoiesis, an IGHV3 gene rearranges toone D and one J element (dotted lines) to form the VH domain. The complementarity-determining region (CDR) 3 is formed by theIGHV3-D-J junction (top). This rearranged gene is first expressed from an upstream promoter as part of an IGHM (a constant μgene inthe H locus), and after a class switch, it is coexpressed with the dedicated IGHG1 genes (with a CH1, hinge, and CH2 and CH3 exons;top of the figure) for assembly in a conventional IgG1 antibody. In other B cells, the IGHV3H is rearranged to one D and one J elementfrom the same locus (dotted lines) to form a VHH domain (below the genome H locus) and eventually coexpressed with an HCAb-dedicatedIGHG2 (or IGHG3) gene (below and to the right of the genome H locus). The CH1 exon region is eliminated during mRNA splicing. TheVH and VHH domains are schematically shown with their framework (FR) and CDRs. The hallmark amino acids for VH and VHH inframework 2 (FR2) are indicated by single letter code and their position in the sequence. (bottom) A folded VHH domain isschematically shown with the A, B, E, and D β-strands of one sheet in the back and the G, F, C, C′, and C′′ strands in the front sheet.The key amino acids of a VH that interact with a VL and that are substituted in a VHH are shown with three-letter code. Thesebecome solvent exposed in a VHH in absence of a VL except for the Phe42 and Gly52, which are generally covered by the long CDR3loop. The CDRs are color coded as in the aligned panel above. Abbreviations: CH, constant domain of the immunoglobulin H chain.



Ann

u. R

ev. B

ioch

em. 2

013.

82. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by Y

ale

Uni

vers

ity -

SO

CIA

L S

CIE

NC

E L

IBR

AR

Y o

n 03

/19/

13. F

or p

erso

nal u

se o

nly.


Single-chain variablefragment (scFv): thepaired VH and VLdomains ofconventionalantibodies tethered viaan oligopeptide

V-NAR: variableantigen-bindingdomain of the newantigen receptorantibodies in shark

is restricted (18). Hence, the loop length andpresence of key residues within the loop predictits actual architecture. In a conventional anti-body, the three HV loops of the VH and thethree HV loops of the VL are juxtaposed andprovide a platform of some 600–900 A2 (theexact interacting surface depends on the size ofthe loop and its amino acids, as well as on thealgorithm used for the calculation). Overall,the paratope of conventional antibodies formsa cavity, groove, or a flat surface (with minorundulations of individual amino acid sidechains sticking out), and these architectureshave been linked to the recognition of smallmolecules, linear peptides, and larger antigens(such as proteins), respectively (19).

Alignment of VHH amino acid sequencesimmediately indicates that the structural orga-nization of the FR and HV regions is similar tothe VH, with a few notable differences in frame-work 2 (FR2) and in CDRs. Within FR2, highlyconserved hydrophobic amino acids (Val47,Gly49, Leu50, Trp52) that normally partici-pate in the interaction with the VL domain (20)are replaced in a VHH by smaller and/or hy-drophilic amino acids (mostly Phe42, Glu49,Arg50, Gly52) (Figure 2) (21–24). Reshapingthis VL side of the domain abrogates any VLinteraction and assists in the nonstickiness ofthe domain in absence of a VL. Indeed, it hasbeen shown that a mouse VH in absence ofthe VL partner is sticky, and the substitutionof FR2 to mimic the amino acids of camelidVHH renders the camelized domain more sol-uble (25). Conversely, the substitution of theamino acids at these positions in a VHH tomimic a human VH leads to the dimerizationof the humanized VHH, where one humanizedVHH structurally replaces the VL domain asin a VH-VL pairing within conventional anti-bodies (26). However, minor deviations wereobserved in the actual dimerization motif (27).Remarkably, in an effort to engineer an auto-nomous V domain of the VH from a single-chain variable fragment (scFv), mutations wereintroduced in the former VL side of the hu-man VH and were screened for the best solution(in terms of expression, monomeric state, and

thermal stability) (28). The amino acid posi-tion and nature of the optimal substitutionswere equivalent but not identical to what na-ture evolved in camelid VHHs. This findingindicated that nature chose one out of a num-ber of possibilities to form an autonomous func-tioning VHH. Although the V-NAR and VHHsequences are very diverse, the increased fre-quency of polar and charged residues at the VLside of the domain supports convergent evo-lution in this region between these two singleantigen-binding domains (29).

The second difference between VH andVHH is observed in the HV loops: Becausethe VHH in an HCAb acts autonomously (in-dependently from the presence of a VL, asseen in conventional antibodies), the antigen isrecognized by only three loops instead of sixloops. To provide a sufficiently large antigen-interacting surface of 600–800 A2 (30), the loopsare longer in a VHH than in a VH of a conven-tional antibody. This was apparent from the se-quence alignments that indicated an enlargedHV region in the H1 loop and the presence ofan extended H3 loop (Figure 2) (31). The en-larged HV loop 1 finds its origin in the germline genes, whereas the longer H3 loop is prob-ably due to the selection of functional VHH do-mains after V-D-J recombination. An extendedloop implies a larger flexibility, and this is ex-pected to be entropically counterproductive forbinding. This issue is solved in camel VHHsby constraining the long loops with a disulfidebond (32). Indeed, many camel VHH sequencescontain an extra pair of Cys residues in the H1and H3 loops, which are known to form an in-terloop disulfide bond (Figure 2), evidently re-stricting the flexibility of the loops in absenceof the antigen. Surprisingly in llama VHH, thisextra disulfide bond occurs less frequently, butthe H3 loop is on average also notably shorterin llama than in dromedary VHH (22–24). Theinterloop disulfide bond in dromedary VHH ismainly formed between the H1 and H3 loops;however, in a small number (∼10%) of VHH, adisulfide bond is also observed between the H3and the FR2 position 50. Also, shark V-NARsevolved to have equivalent disulfide bonds

17.6 Muyldermans


Ann

u. R

ev. B

ioch

em. 2

013.

82. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by Y

ale

Uni

vers

ity -

SO

CIA

L S

CIE

NC

E L

IBR

AR

Y o

n 03

/19/

13. F

or p

erso

nal u

se o

nly.


tethering the antigen-binding loops (33, 34).In llama VHHs, and sporadically in dromedaryVHHs as well, we noticed an interloop disul-fide bond between the H3 loop and the H2 loop(position 55) (22, 23). Although the position ofthe Cys in the H1 loop is restricted at positions30, 32, or 33 (position 50 in FR2, or position55 in H2), in the H3 loop, the Cys occurs atnearly all possible positions in the N end, themiddle, or the C end of the H3 loop (14). In alimited number of VHHs, the H3 loop had anextended conformation stretching out from theremainder of the domain (as in the VH of a con-ventional antibody). This was observed mainlywhen the VHH had a Tyr at position 42 anda shorter than usual H3 loop length as in agreen fluorescent protein (GFP) binder (35). InVHH with Phe42, the long H3 loop adopts aspecific conformation, named stretched twistedturn, whereby the base of the loop is twisted,and the tip is folded toward the C′ strand (the7th or 8th amino acid counting from the C endof the H3 loop approaches the FR2 amino acidat position 52, especially when it is Gly) (36).In this H3 configuration, the Phe42 is shieldedfrom contact with water by the H3 loop(Figure 2). An aromatic core formed by Trp118(FR4), Tyr93 (FR3), and Phe117 or Tyr117probably stablizes this H3 loop conformation.

The prolate (rugby ball–shaped) structure ofthe VHH domain forms a convex paratope sur-face, which makes it extremely suitable to insertin cavities on the surface of the antigen (30).This type of interaction increases the actualinteraction surface of the paratope. The longH3 sequences can form a protruding loop [asin cAb-Lys3 (37)]; however, this seems to bea special case, and the long loop in most Nbstructures folds over the FR2 region to form aflat paratope surface (38). As in V-NAR wherethe HV4 region (loop between the D-E strands)participates actively in antigen recognition, thisloop in a VHH is also often close to the antigen.Although there is some increased variability atthis position in the domain, its involvement inantigen binding is rather limited. In contrast,the amino acids at the N end of the extendedH1 loop are regularly highly involved in anti-

gen recognition (39). Overall, the paratope ar-chitecture forms a variety of structures rang-ing from protruding loops shared with V-NARs(33, 37) to a flat surface (38), and also a cavityas seen for the antihapten binders (40, 41).

DEDICATED HEAVY-CHAINANTIBODY GENES

Organization of the H Locus in theGenome of Camelids

The human and mouse H gene locus comprisesa sequential organization of multiple V ele-ments, D elements, and JH elements upstreamof the constant immunoglobulin genes IGHM,IGHD, IGHG, IGHE, and IGHA, although avariable copy number for each gene per haploidcan be found in different animals (42). Like-wise, the elements that produce the H chain forconventional antibodies and for HCAb are lo-cated in the same locus in the camelid genome(Figure 2) (8, 43). It is clear that dromedaryand the alpaca employ dedicated IGHGgenes to produce HCAbs. Thus, some IGHGgenes are used exclusively for conventionalantibodies, and a different set of IGHGgenes will produce HCAbs (44, 45). Comparedto the IGHG genes for conventional antibod-ies, these genes have an identical organizationof exons, introns, switch regions, and alter-native polyadenylation sites that produce se-creted and membrane-bound IgG molecules.Nucleotide sequences for the CH1 exon areembedded; however, they carry a nucleotide Gto a point mutation that disrupts the consen-sus splicing site (GT) at the 5′ end of the in-tron between the CH1-hinge exons and thatprovokes the elimination of the CH1 regionfrom the mRNA by splicing. Most likely, thereare no camel-specific factors required for thisunique splicing, as the dromedary IGHG2agene, complemented with an upstream pro-moter and rearranged VHH-D-J gene fragmentand introduced in an NSO mouse cell lineor in a transgenic mouse, is properly splicedand generates functional HCAbs in the cul-ture supernatant or serum, respectively (46, 47).



Ann

u. R

ev. B

ioch

em. 2

013.

82. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by Y

ale

Uni

vers

ity -

SO

CIA

L S

CIE

NC

E L

IBR

AR

Y o

n 03

/19/

13. F

or p

erso

nal u

se o

nly.


In contrast, a transgenic mouse with thedromedary-specific CH1 mutation in the hu-man IGHG2 and IGHG3 genes fails to pro-duce functional HCAbs (48). Successful HCAbscould be generated if the entire CH1 exonwas deleted from the same genes. This in-dicates that (unidentified) surrounding nu-cleotides of the CH1 exon/intron splicing sitein the camelid IGHG genes might be essentialfor proper VHH-hinge splicing. The analysisof dromedary genomic clones further identifiedthe existence of putative IGHG genes for HCAbthat were never observed in cDNA sequencesand that probably reflect pseudogenes (29).

It seems that the alpaca genome containsonly one IGHM gene with a functional CH1splicing site (43). Also, the cDNA analysisshowed that the mRNA with a rearranged VH-D-J and that with a VHH-D-J region maintainsthe CH1 exon sequences.

Dedicated VHH Genes in the H Locus

The camelid genome also harbors dedicatedVHH germ line genes (denoted as IGHVH)interspersed with the VH germ line genes forconventional antibodies (designated as IGHV)(Figure 2) (43, 49). The IGHV and IGHVHgenes are easily distinguished because theyencode hallmark amino acids of FR2. Thedromedary genome contains about 50 IGHVand 40 IGHVH genes (31), whereas the alpacaencodes 71 IGHV and 17 IGHVH genes (43).During B-cell lymphopoiesis, either an IGHVor IGHVH gene is involved in the V-D-Jrearrangement to form the VH or the VHHdomain, respectively. Because the D and J genepool are common for both the VHs and VHHs(Figure 2), it might come as a surprise that theH3 loops of VHHs (formed by the V-D-J junc-tion) are on average longer than those of VHs.The longer CDR3 in VHH domains might beexplained in several ways: (a) a higher activityof deoxynucleotidyl transferase during theV-D-J rearrangement, but this is less plausibleas it implies that a distinct subset of B cells ispredestined to produce HCAbs; (b) a selectionat the pre-B-cell receptor stage, a checkpoint to

ensure that the V-D-J rearrangement producesa properly folded V domain; or (c) a selection onfunctionality where it is hypothesized that thoseVHH domains with shorter H3 loops will mostlikely have a too small putative antigen-bindingsurface, which makes it less likely that such Bcells will bind to antigen and be expanded dur-ing an immune response. Interestingly, IGHVHgenes also differ from IGHV genes because ofthe presence of a Cys codon in the CDR1 region(or position 50 in the FR2 region) in dromedary(31) or Cys55 in alpaca VHH (43). The secondCys codon in the H3 loop is introduced duringthe V-D-J recombination or afterward bysomatic hypermutation. Obviously, a B cellwhere the V-D-J recombination productcontains only one single Cys (apart from theconserved Cys23-Cys94 pair) will be counters-elected if it does not manage to replace this Cysby mutating the Cys codon or to introduce asecond Cys codon at a suitable position to forma cystine. Indeed the presence of one singleCys in the paratope might lead to VHH dimers(50) whereby antigen binding is obstructed.

The IGHVH genes differ also from IGHVgenes in a more subtle way. Apparently,the IGHVH genes—but not the IGHVgenes—often contain peculiar sequences (i.e.,palindromes or heptamer-like sequences ofthe Ig recombination signal sequence) that arerecognized as unstable or prone to unequalDNA recombination, gene conversion, or genereplacement events (31). It has been hypothe-sized that such elements would lead to a fastergene evolution and thus to a faster expansionof the IGHVH repertoire (29). In addition,these elements might also be responsible forB-cell receptor editing (by a gene conversionmechanism with an IGHVH gene located up-stream). Indeed, the cDNA sequence analysisof VHH with a known antigen specificity andretrieved after phage display very often revealsthe sequences with an identical H3 loop (sameV-D-J rearrangement) and with different H1and/or H2 loop sequences, although it cannotbe excluded that these binders were producedartificially by polymerase chain reaction (PCR)crossover.

17.8 Muyldermans


Ann

u. R

ev. B

ioch

em. 2

013.

82. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by Y

ale

Uni

vers

ity -

SO

CIA

L S

CIE

NC

E L

IBR

AR

Y o

n 03

/19/

13. F

or p

erso

nal u

se o

nly.


GLEW motif:a hallmark Gly49,Leu50, Glu51, Trp52motif, conserved in theframework 2 region ofall VH domains

Another surprising and subtle VHH se-quence imprint has been introduced in the re-gion encoding the H1 loop (31). In human andmouse VH sequences, this loop, bridging thetwo β-sheets of the V domain, contains con-served amino acids at positions 25, 27, 28, 30,and 39. These amino acids (Ala, Gly, Phe, Phe,and Met, respectively) are considered to be thekey amino acids for the type 1 canonical loopstructure (51). In the VHH germ line genes,they are conserved with one notable difference,that is, the Phe28 and Ph30 (TTT and TTCcodons) were replaced by Tyr codons (TATand TAC). It has been reported that TAY basesteps are hot spots for somatic hypermutation(i.e., are more prone to diversify), and indeed,these codons are very often mutated in affinity-matured VHH domains (31, 43). In addition,this codon mutagenesis is the main contributorto the extension of the HV region in the H1loop toward the N-terminal loop region, andthe H1 loop structure often deviates from theknown canonical structures observed in humanor mouse VH (52).

Promiscuous VH Genes Contribute tothe VHH Repertoire

An antigen-binding site formed by two do-mains, which are first assembled independentlyby V-(D-)J gene recombinations and then com-bined, leads to a seemingly unlimited numberof possible paratope structures and thus to a fit-ter immune system based on a minimal numberof gene elements. It is surprising that HCAbsthat lack the VH-VL combinatorial diversifica-tion can compete with conventional antibodies.The structural repertoire of the VHH paratopewas even more restricted than originally antici-pated when it was shown that two VHHs raisedin two different dromedaries (in Morocco andin the United Arab Emirates immunized withthe same antigen at different times) were as-sembled from different IGHVH genes and dif-ferent J minigenes but from the same D ele-ment (53). In addition, it was shown that theloop region encoded by the D element in thetwo VHHs adopted the same structure and in-

teracted with the same epitope. However, theJ and V gene elements apparently underwentconvergent somatic hypermutations. This find-ing underscores the strength and importanceof somatic hypermutations and selection dur-ing the affinity maturation of these antibodies.

The number of available VH (or VL)families is another contributor to possibleparatope diversity. Up to a few years ago, allIGHVHs were reported to belong to IGHV3family, the most abundant and widespreadfamily in human and mouse genomes, andused for most versatile, functional antibodies(18). However, recently the occurrence of VHgerm lines of IGHV1 and of IGHV4 (clan II)were identified in alpaca (43). These geneswere reported to produce a VH domain forconventional antibodies. The IGHV4 genesare also present in dromedary, where these Velements are surprisingly promiscuous as theycan produce VH domains and VHH domains,the autonomous domains that function inantigen binding in HCAbs in absence of a VLpartner (54). These VH-like single domainswith an IGHV4 imprint had a slightly lowerhydrophobicity profile in their FR3.

Likewise, the IGHV3 genes (with a hy-drophobic GLEW motif in FR2) are also foundto produce HCAbs (without a CH1 domain).Therefore, the IGHV3 genes with a GLEWmotif in camelids might be promiscuous as welland might function after V-D-J rearrangementand expression either paired with a VL in aconventional antibody or autonomously inan HCAb. Indeed, multiple antigen-bindingVH-like domains with a GLEW motif havebeen isolated repeatedly from HCAb libraries.Sometimes, these VH-like domains clearly havea sequence imprint that prevents an associationwith a possible VL as they lack the conservedTrp118. The Trp118 is encoded by the J ele-ment (Figure 2), is 100% conserved in humanand mouse VHs, and is strongly involved inthe VH-VL interaction (20). However, duringthe D-J rearrangement in camelids and theconcomitant nucleotide insertion/deletionprocess, a deletion of the 5′ J sequences mighttruncate the Trp codon (TGG) so that this



Ann

u. R

ev. B

ioch

em. 2

013.

82. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by Y

ale

Uni

vers

ity -

SO

CIA

L S

CIE

NC

E L

IBR

AR

Y o

n 03

/19/

13. F

or p

erso

nal u

se o

nly.


Class switch: therearrangement of aV-D-J exon from the5′end of the IGHM toan IGH gene locateddownstream

Pre-B cell: a B cell atan early stage ofdevelopmentexpressing the heavychain associated with asurrogate light chain

Surrogate light (L)chain: a substitute ofthe classic L chainwhere separate VpreBand λ5 polypeptidesreplace VL and CLdomains

codon is changed into an AGG, CGG, or GGGcodon (if only T is deleted). In two out ofthree of these cases, Trp is substituted by Arg(once by Gly), and this mutation is regularlyobserved in the camelid VHHs (belonging tothe HCAbs). Independently of whether a VHor VHH germ line was employed, the presenceof an Arg (rather than Trp) will undoubtedlyprevent proper association with a VL. Inhuman or mice (or any other mammal, exceptcamelids) that do not have HCAb IGHG genes,such VHs with Arg118 are never observedbecause B cells with this mutation probablycannot survive during B-cell lymphopoiesis.

Heavy-Chain Antibody Ontogeny

The occurrence of an IgM stage in an HCAbformat is another enigma. Actually, we do notknow how the class switch from an IgM to IgGoccurs if an IGHVH gene is rearranged in thepre-B cell. Our best guess, and current work-ing hypothesis, is that a rearranged IGHVH-D-J (or an IGHV-D-J∗ where J∗ indicates that theconserved Trp118 codon has been modified)might fail to be expressed on the surface of thepre-B cell in combination with the surrogateL chain, a dimer of VpreB and λ5 (55). Nor-mally, in the pre-B cell, it is suggested that afterthe IGHVH-D-J rearrangement, the unfoldedCH1 of the VHH-IGMH translation productassociates with the BiP protein and is retained inthe endoplasmic reticulum (56).This BiP pro-tein should be exchanged by the surrogate Lchain (where λ5 functions as a CL and VpreBas a VL) before the L chain has been rear-ranged for surface expression of the H chainand for pre-B-cell receptor signaling. Failure ofthe VpreB to associate with the VH domain willinterfere with proper exposure of the pre-B-cellreceptor and its signaling, causing the B cells torapidly disappear (57), although escape routesmight be in place. Where cells blocked in thesurface expression of the pre-B-cell receptor arenormally ablated, in camelids they might be res-cued by an immediate class switch from IGHMto one of the dedicated IGHG genes for HCAbs.Indications in favor of such an ontogeny comes

from (a) the fact that transcripts of IGHVH-D-J joined with an IGHM are always observedwith an intact CH1 exon, (b) the extremely lowproportion of IGHVH-D-J (less than 5%) com-pared to IGHV-D-J with IGHM, and (c) theabsence of somatic mutations in IGHVH whencoexpressed with IGHM (43).

BIOCHEMICAL PROPERTIESOF NANOBODIES

Identification of Antigen-SpecificNanobodies

Nbs, the recombinant single variable domainsderived from camelid HCAbs, are generallyreadily obtained after a brief immunization,followed by cloning the V gene repertoire fromperipheral blood lymphocytes and by selectionthrough phage display (58). Antigen-specificHCAbs are affinity matured during a shortimmunization step, mostly with proteins as theimmunogen but also by a DNA-prime, protein-boost strategy (59). Because the entire antigen-binding fragment of the HCAb consists of oneVHH domain, encoded by a gene fragmentof only ∼360 bp that is easily amplified byPCR in one single amplicon, small librariesof ∼106 individual transformants are alreadyrepresentative of the immune VHH repertoireof B cells present in a blood sample of ∼50 ml.In cases where immunization is not practical(i.e., no available, toxic, or nonimmunogenicmolecules), immune VHH libraries can be sub-stituted by naive, semisynthetic, or syntheticV repertoires (60, 61), but such libraries needto be much larger (∼109 individual clones) toallow the retrieval of high-affinity binders.

Antigen-specific Nbs are retrieved from im-mune or other libraries by phage display orany other selection protocol, including bacterialdisplay, yeast display, intracellular 2 hybrid se-lection (62, 63), ribosome display (64), and oth-ers. However, because of the robustness of thetechnology, the cloned V library is expressedpreferably on a phage and panned on an anti-gen that is immobilized in wells of microtiterplates by passive adsorption, or when the

17.10 Muyldermans


Ann

u. R

ev. B

ioch

em. 2

013.

82. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by Y

ale

Uni

vers

ity -

SO

CIA

L S

CIE

NC

E L

IBR

AR

Y o

n 03

/19/

13. F

or p

erso

nal u

se o

nly.


antigen is biotinylated on streptavidin-coatedsolid supports (65). Normally, two to threerounds of panning are sufficient to enrich theclones so that individual clones can be screenedfor production of antigen-specific Nbs in astandard enzyme-linked immunosorbent assay(ELISA). Nucleotide sequencing of the ELISA-positive clones is used to deduce the amino acidsequence of the Nb. The entire procedure toidentify useful antigen binders is fast as multipleimmunogens or even proteome fractions can beused to immunize one animal in less than twomonths (66, 67). In this way, the immune librarycan be used to select binders to various anti-gens in parallel. Thus, Nbs against hundredsof antigens can be identified each year by oneresearcher.

Affinity Parameters of Nanobodies

In contrast to the cloning of a scFv, wherebythe VH and VL pairs that were affinity ma-tured in B cells became scrambled by randomassembly of the VH exons and VL exons thatwere first individually PCR amplified, thePCR amplification and cloning of the VHHpresent in one exon allowed the cloning ofintact, affinity-matured VHH from peripheralB lymphocytes as a whole. Therefore, theantigen specificity and affinity of Nbs fromimmune libraries are of good quality. Kinetickon and koff rate constants in the ranges of 105 to106 M−1s−1 and 10−2 to 10−4 s−1, respectively,are routinely obtained so that low nanomolar oreven picomolar equilibrium dissociation con-stants are obtained. Such affinity parametersare excellent for most applications. Neverthe-less, in vitro affinity maturation approaches,such as error-prone PCR, spiked mutagenesiscombined with ribosome display (68) andAla scanning-based mutations to identify thecritical amino acids for antigen recognitionhave been successfully introduced to improvethe stability of the domain and/or the affinityfor their cognate antigen (69). In an alternativeapproach, a few carefully selected mutationsat the edge of the paratope were introducedto measure their effect on antigen-Nb kinetic

and equilibrium affinity values. These datawere combined with a multivariate analysis ofthe parameterized quantitative descriptorsof the mutations and buffers to propose aquantitative predictive algorithm that modelsthe affinity parameters of all other possiblemutants at those positions (70).

Expression of RecombinantNanobodies

The Nbs are expressed to a high level inmicroorganisms (71–73), mammalian celllines, and plants (74). For bacterial expression,cloning the Nb after a secretion signal is pre-ferred so that it is produced in the periplasm,where the oxidizing environment formsdisulfide bonds properly. The purification isstraightforward because Nbs are usually clonedin frame with an His6 tag so that, after periplas-mic extraction (which already avoids majorcontaminations with cytoplasmic proteins),immobilized metal affinity chromatographyand a gel filtration produce pure Nbs. Yieldsof several milligrams per liter of culture areroutinely obtained from simple culture flasks.For applications where Nbs without any tagsare required, it is recommended to select Nbsthat bind to protein A. Most VHHs are closelyrelated to members of human VH family 3,and such proteins are often retained on thisaffinity adsorbent (75).

Stability of Nanobodies

Nbs are easily concentrated by ultrafiltration(with filters of mol wt cutoff <5,000) to1–10 mg/ml−1 in standard phosphate or Trisbuffers for usage as a stock solution. The shelflife of Nbs is excellent; they are stored formonths at 4◦C, and even longer at −20◦C whilemaintaining full antigen-binding capacity. In-cubations at 37◦C for several weeks seem to betolerated as well (76). Moreover, Nbs are robustunder stringent conditions and resist chemicaland thermal denaturation. Usually Nbs dena-ture ∼2.3–3.3 M guanidinium at ∼60–80◦C(77). Some Nbs express an unusually stable



Ann

u. R

ev. B

ioch

em. 2

013.

82. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by Y

ale

Uni

vers

ity -

SO

CIA

L S

CIE

NC

E L

IBR

AR

Y o

n 03

/19/

13. F

or p

erso

nal u

se o

nly.


behavior, resist temperatures above 90◦C (78),and have a chemical stability of 60 kJ/mol−1 (62,77). The stability of a Nb can be increased byintroducing Cys at position 54 and 78 (on theC′′ and E strands, respectively) to form an extradisulfide bond (79, 80). These disulfide bond–stabilized Nbs turn out to be highly resistantto degradation by pepsin or chymotrypsin (81),suggesting that such Nbs can be administeredorally. Trypsin-resistant Nbs for oral deliveryhave been obtained after random mutagenesisthrough DNA shuffling and selection fortrypsin-resistant variants (82). Furthermore,pannings with phage-displayed VHHs havebeen performed under harsh conditions (e.g.,in presence of detergents or denaturants),and there were indications that such selectedVHHs are also resistant to other denaturingconditions, such as high urea concentrations(83). The high-temperature and acid toler-ance of the filamentous phage has also beenexploited to select for sdAbs with enhancedresistance or reversible refolding capacity uponreturning to physiological conditions (84, 85).

Because the Nb sequence (VHH family3) is rather conserved, the scaffold of somehighly stable Nbs have been used successfullyto graft the antigen-binding loops of less-stableantigen-specific Nbs to obtain a chimeric Nbwith the stability of the loop acceptor Nb andthe antigen-specificity of the loop donor Nb(86). However, this strategy is not always real-istic, as much of the Nb stability is dictated bythe CDR3 sequence (87). Therefore, graftingCDR3 loops will also transfer their intrinsiccontribution to stability. In addition, inasmuchas the CDR3 amino acids are either in directcontact with the antigen or they maintainand influence the conformation of the CDR3amino acids that directly contact the antigen,the CDR3 amino acids responsible for reducedstability cannot be replaced without seriousloss of affinity (88).

Nanobodies Are Nonimmunogenic

The small size, stable behavior, rapid clearancefrom blood, and a sequence sharing a high

degree of identity with human VH are allproperties that predict low immunogenicity ofa Nb. Indeed, no immune response against theNb moiety was raised in mice or humans thatwere injected with Nb-containing constructs(89–91). Moreover, a strategy was developedto humanize 12 out of the 14 amino acids thatdiffer between human VH and camelid VHH(92).

Recognition of Unique Epiopes byNanobodies and Multivalent and/orMultispecific Constructs

The unique (convex) paratope architecture ofsdAbs prefers to associate with concave sur-faces of the antigen of sdAbs, such as the activesite of an enzyme. Therefore, it is no surprisethat many Nbs affect the catalytic activity ofenzymes (10, 93–95). Where some of the Nbsclearly behave as allosteric inhibitors (96, 97),others might activate enzymatic activity (98)possibly as they stabilize the enzyme in an openconformation.

The strict monomeric behavior of the Nbin combination with its minimal size makes itan ideal building block to develop multidomainconstructs (99). Several dimeric Nb constructshave been identified: (a) the bivalent, monospe-cific construct whereby the product of tandemlycloned, identical Nbs increases the functionalaffinity (avidity) for the antigen; (b) the bi-paratopic, monospecific construct obtainedby tandem cloning of two Nbs recognizingdifferent epitopes on the same antigen, whichpossibly chelates the antigen and concomitantlybinds with higher avidity (100); and (c) thebispecific construct generated by formation of atandem pair of Nbs with each recognizing a dif-ferent antigen, a process used to tether two in-dependent antigens (101, 102). In an alternativeapproach, the Nb is cloned to a protein moietywith a natural tendency to dimerize or to mul-timerize, such as a leucine zipper motif or thepentamerizing motif of verotoxin (103, 104).Furthermore, the antigen-specific Nb gene hasbeen cloned in frame with the hinge, CH2 andCH3 exons of human, and mouse or pig IgG

17.12 Muyldermans


Ann

u. R

ev. B

ioch

em. 2

013.

82. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by Y

ale

Uni

vers

ity -

SO

CIA

L S

CIE

NC

E L

IBR

AR

Y o

n 03

/19/

13. F

or p

erso

nal u

se o

nly.


and subsequently transfected in myeloma celllines to reconstruct chimeric HCAbs (104, 105).

APPLICATIONS WITHNANOBODIES

A renewable (or sustainable) source, economicproduction, small size, human(-ized) sequence,stable and soluble behavior in aqueous solu-tions, reversible refolding, and specific andhigh affinity for only one cognate target arethe ideal properties for a practical binder. Nor-mally, the Nbs selected from immune librariesmeet all these requirements without the needfor additional improvements. The notions thatNbs might recognize epitopes that are notantigenic for conventional antibodies and thatmany of them modulate the function of thetarget are supplementary assets. Obviously,these beneficial properties stimulated severalresearch groups in universities and pharmaceu-tical and biotech companies to employ Nbs as aresearch tool and/or to develop future diagnos-tic and therapeutic applications. This resultedin numerous applications whereby the proof ofprinciple has been successfully demonstrated;the applications have been covered recentlyelsewhere (106–108). Here, we focus on a num-ber of examples wherein Nbs offer a specialadvantage over other equivalent binders andwhere Nbs surpass the “me-too” qualification.

Nanobodies as Research Tools

The fast retrieval of high-affinity binders to var-ious targets with the Nb identification technol-ogy permits the assembly of a comprehensiveresource of well-characterized reagents for im-portant targets, including intracellular signal-ing molecules and cancer biomarkers, as versa-tile tools in this postgenomic era (109).

The genetic fusion of a fluorescent proteinwith a Nb, and its intracellular expression,produces useful chromobodies or fluobodies totrace the antigen in various cellular compart-ments in living cells (110, 111). Moreover, theuse of GFP-binding Nbs coupled to organicdyes on any GFP-tagged construct enabled

single-molecule localization with superresolu-tion imaging techniques (112). The specificityof a Nb for particular conformational variantsof its target and its capacity to induce andreport conformational changes have beenexploited for monitoring by ratio imagingtarget protein expression, its translocation, andfinal subcellular localization (35, 113–115). Inparallel research, it has been confirmed thatintracellularly expressed sdAbs remain solubleand exhibit specific antigen recognition activ-ities to abrogate particular protein functions ofthe antigen inside cells. The identification ofsuch sdAbs, which compete with normal in situprotein-protein interactions, provides a toolfor target validation and a lead molecule toinvestigate difficult interactions or interactionsconsidered undruggable (116, 117).

Apart from blocking a surface of the targetprotein to interfere with proper protein-proteininteractions, it is also possible to trigger thedepletion of antigen via the ubiquitin pathway.Proof of this concept was given in Drosophilamelanogaster and Danio rerio models by intracel-lular expression of the GFP-specific Nbs fusedto the F-box domain to recruit polyubiquitina-tion machinery and to initiate the proteasome-mediated degradation of the target, capturedin the ubiquitinated Nb complex (118).

Several years ago, antibodies were pro-posed to assist the crystallization process andstructural determination of flexible or aggre-gating ‘ “high-hanging fruit” proteins. Apartfrom a few isolated cases, it is only now thatrapid progress has been made because of theavailability of recombinant single-domain affin-ity reagents. These affinity reagents fix highlydynamic proteins in a binder-preferred con-formation and stabilize the intrinsic flexibleregions or detergent-solubilized membraneproteins by shielding hydrophobic antigen sur-faces from contact with solvent to allow theformation of effective crystal contacts (119).Again, it is the fast identification of antigen-specific Nbs as well as the high-productionyields, small size, robust structure and targetingclefts on the surface of the antigen (which oftencoincide with active enzymatic sites or



Ann

u. R

ev. B

ioch

em. 2

013.

82. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by Y

ale

Uni

vers

ity -

SO

CIA

L S

CIE

NC

E L

IBR

AR

Y o

n 03

/19/

13. F

or p

erso

nal u

se o

nly.


ligand- or receptor-binding cavities) that al-low Nbs to tackle the crystallization and struc-tural determination of these challenging targets(120–123).

Nanobodies as Diagnostic Tools

The development of Nbs for the quantitativedetection or the simple capture of their targetshas been surprisingly slow. The larger size ofthe conventional antibodies is probably bettersuited for random coupling to solid surfaces.However, the availability of high-affinity Nbsthat are easily modified to avoid chemically re-active groups (primarily amines from lysines)in the vicinity of the paratope and the inclusionof such groups at the opposite end of the do-main that permit their directional immobiliza-tion on the sensor surface for the maximal cap-turing capacity of the antigen. This allowed theuse of Nbs to generate sensitive and selectivebiosensors (124).

In another setting, Nbs were conjugatedchemically to branched gold nanoparticles toproduce an effective antigen-targeting pho-tothermal therapeutic upon irradiation withlaser light in the near biological window (125).

The fusion of the Nb gene with the genefor the magnetosome protein MamC andexpression in the magnetite-synthesizingMagnetospirillum gryphiswaldense bacteria offeran elegant strategy for producing Nbs directlycoupled to magnetic nanoparticles (126) to cap-ture and enrich analyte at low concentrationsin complex mixtures. The covalent linkage ofNbs to a solid, inert, or magnetic support seemsto be a highly valued approach to generateaffinity adsorbents. This was shown by devel-oping the nanotrap, an immobilized Nb thatrecognizes the GFP as a target (115), and byNbs directed against various immunoglobulinisotypes (127). Such affinity adsorbents havebeen commercialized by ChromoTek GMbH(Planegg-Martinsried, Germany) and BAC BV(Naarden, Netherlands), respectively. The lat-ter company recently introduced a tag-bindingNb to their portfolio. This Nb specificallyrecognizes the C-terminal tetraamino acids

Glu-Pro-Glu-Ala (EPEA) (128) that can becloned as a tag behind any protein. The smallsize of Nbs means that few proteins from acomplex mixture in an extract will recognizethe sdAb, and this significantly reduces thenonspecific adsorption of molecules.

Owing to their small size, which is wellbelow the renal clearance cutoff of mol wt∼50,000, Nbs are rapidly eliminated fromblood, and this is exactly what is required fora good in vivo–imaging agent (129). Rapidtargeting to diseased tissue and fast bloodclearance make use of short-lived nuclides(68Ga, 18F) with half-life times (t1/2) of 68 and110 min, respectively, which make positronemission tomography and computed tomogra-phy imaging practical for measuring picomolarconcentrations of the tracer within 1–3 hpostinjection and thus result in a very low radi-ation burden for the patient. This was recentlytested in mice xenografts with Nbs against anepidermal growth factor receptor labeled with68Ga, yielding a tumor to blood ratio of 25after 3 h (130) against HER-2, a breast cancerantigen (131), and V-CAM1, an antigen usedto diagnose vulnerable atherosclerotic plaques(132).

Nanobodies as Therapeutics

Antibodies have been used for ages for passiveimmunization to treat envenomed victims orinfected patients. Although polyclonal anti-bodies generally result in better neutralizationand thus better protection, a monoclonal Nbscreened to recognize special epitopes involvedin receptor recognition can reach an extremelyhigh-neutralization potency (133). In additionto scorpion toxins, several antibacterial toxins(134) and antisnake venom Nbs are actively be-ing investigated (135–137). Moreover, in somecases, the large size of conventional antibodiesobstructs the access of hidden and essential epi-topes on pathogenic agents, such as viruses, bac-teria, or parasites; in such cases, Nbs will havespecial advantages as therapeutics (138–146).

Several other Nb-derived therapeutics(e.g., anti-IL6R or anti-TNFα used to treat

17.14 Muyldermans


Ann

u. R

ev. B

ioch

em. 2

013.

82. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by Y

ale

Uni

vers

ity -

SO

CIA

L S

CIE

NC

E L

IBR

AR

Y o

n 03

/19/

13. F

or p

erso

nal u

se o

nly.


inflammatory diseases, an anti-von Wille-brand factor used to treat patients with acutethrombotic thrombocytopenic purpura, aswell as an anti-RANKL used to combat boneloss disorders, and others) are in the pipeline;some have passed phase I and phase II tests(147). However, the competition from othertherapeutic compounds is huge. The successof Nbs in therapy will come from a morepatient-friendly administration (topical, oral,or inhalation) and from obtaining improvedblood levels over a prolonged time, as well as byfine-tuning the residence time, by linking theNb to albumin-binding moieties, or by chang-ing their hydrodynamic volume in various ways(148).

In addition, the added value of affinity-matured Nbs originates from their capacity todiscriminate the cognate target from closelyrelated variants; such high specificity oftencannot be achieved by most small organicantagonists. Therefore, Nbs have been ex-ploited to block unwanted enzymatic activities,e.g., the cytotoxic effects of T-cell ecto-ADP-ribosyltransferase ART2.2 in lymphatic organs(59).

Finally, Nbs have also been assessed in theircapacity to deliver cargoes to tissues that aredifficult to access. To this end, Nbs were gen-erated that cross the blood-brain barrier or thatlead to transcytosis across epithelia (100, 149–151).

SUMMARY POINTS

1. An accessible and streamlined protocol has been developed for the fast generation ofseveral Nbs that recognize with high specificity and high affinity their cognate antigen.Recombinant Nbs are well expressed in a variety of microorganisms, highly robust, andeasy to engineer according to the needs of the user or for optimal performance in theenvisaged application.

2. Multiple crystal structures of Nbs in complex with their antigen are available. Theseprovide detailed insight that supports rationally designed engineering to increase thepotency of Nbs.

3. It is well established that Nbs prefer to interact with cavities on the surface of theirantigen, such as the catalytic site of enzymes or the ligand-binding site of receptors.Therefore, many Nbs exert an inhibiting, agonistic, or antagonistic effect on theirantigen.

4. Targets that switch rapidly between different conformations can be trapped and stabilizedin the conformation that is preferred by Nbs. This leads to the identification of unknownconformations and allows detailed investigation of, for example, enzyme mechanisms.Nbs might stabilize otherwise unstable proteins and prevent amyloidosis.

5. The strict monomeric behavior of Nbs makes them a perfect tool to include in largerconstructs where the Nb provides antigen specificity and the targeting function.

6. Nbs seem successful as crystallization chaperons. Several membrane proteins and unsta-ble proteins that are prone to degradation, unfolding, or aggregation have been stabilizedwith Nbs, and an impressive number of crystal structures of such high-hanging fruit pro-teins have been determined in the past years.

7. The availability of techniques for radiolabeling Nbs in conjunction with their small size,which is well below the renal cutoff, makes them ideal tools for noninvasive in vivoimaging.



Ann

u. R

ev. B

ioch

em. 2

013.

82. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by Y

ale

Uni

vers

ity -

SO

CIA

L S

CIE

NC

E L

IBR

AR

Y o

n 03

/19/

13. F

or p

erso

nal u

se o

nly.


FUTURE ISSUES

1. Conventional antibodies and HCAbs are elicited upon immunization with antigens.However, occasionally the response in the HCAb is disappointingly low (for example,small peptides and haptens). Therefore, better knowledge about the effects of adjuvantson the immune system of camelids is necessary. Moreover, developing a clear picture ofthe different HCAb isotypes (subisotypes and allotypes) occurring in sera and the exactfunction and role of each of these subisotypes is high priority.

2. The ontogeny of functional HCAbs in B cells of camelids remains a black box. Althoughthe majority of genes have been identified, we do not understand how and why theV-D-J rearrangement involving an IGHV3H gene has a longer CDR3 or how and whenthe class switch from IGHM to IGHG2 or IGHG3 occurs. In addition, any informationon the involvement and role of the surrogate L chain at the pre-B-cell stage would bewelcome.

3. The toolbox of Nb-based products needs further development for target validationand to facilitate cutting-edge science. We expect additional developments for Nbsas highly specific capturing agents; in chromatin immunoprecipitations; in intracel-lular imaging; and in specific degradation, retention, or translocation of intracellulartargets.

4. Investigators need to implement Nbs in the development of phenotypic selection strate-gies to obtain target-modulating Nbs.

5. It is expected that Nbs will be generated that create access to unique locations, such asthose that cross the blood-brain barrier or that could deliver a cargo to particular cells andpenetrate their cell membrane. Some of these objectives could be reached by equippinglentiviral or adenoviral vectors with Nbs.

6. The development of a Nb in a next-generation therapeutic and its commercializationare anticipated.

DISCLOSURE STATEMENT

The author is not aware of any affiliations, memberships, funding, or financial holdings that mightbe perceived as affecting the objectivity of this review.

ACKNOWLEDGMENTS

I thank all researchers from our laboratory and collaborators all over the world for sharing theirexciting data and for their well-appreciated contributions in the past 20 years, thereby formingthe basis for the success of Nb research. Apologizes are offered to those whose work could not beappropriately cited owing to space restrictions. I extend special thanks to all granting authoritiesfrom the EU (especially the AFFINOMICS Project 241481), NATO (Science for Peace),Instituut voor Aanmoediging van Innovatie door Wetenschap en Technologie in Vlaanderen(IWT) and Fonds voor Wetenschappelijk Onderzoek-Vlaanderen (FWO), VIB funding, andOnderzoeksraad (OZR) of Vrije Universiteit Brussel for their financial support of our Nbresearch.

17.16 Muyldermans


Ann

u. R

ev. B

ioch

em. 2

013.

82. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by Y

ale

Uni

vers

ity -

SO

CIA

L S

CIE

NC

E L

IBR

AR

Y o

n 03

/19/

13. F

or p

erso

nal u

se o

nly.


LITERATURE CITED

1. Padlan EA. 1994. Anatomy of the antibody molecule. Mol. Immunol. 31:169–2172. This is the firstdescription of theoccurrence of functionalheavy-chain-onlyantibodies in sera ofcamelids.

2. Hamers-Casterman C, Atarhouch T, Muyldermans S, Robinson G, Hamers C, et al. 1993.Naturally occurring antibodies devoid of light chains. Nature 363:446–48

3. Alexander A, Steinmetz M, Barritault D, Frangione B, Franklin EC, et al. 1982. γ Heavy chain diseasein man: cDNA sequence supports partial gene deletion model. Proc. Natl. Acad. Sci. USA 79:3260–64

4. Cogne M, Preudhomme JL, Guglielmi P. 1989. Immunoglobulin gene alterations in human heavy-chaindiseases. Res. Immunol. 140:487–502

5. Morrison SL. 1978. Murine heavy-chain disease. Eur. J. Immunol. 8:194–996. Flajnik MF, Kasahara M. 2010. Origin and evolution of the adaptive immune system: genetic events and

selective pressures. Nat. Rev. Genet. 11:47–597. Flajnik MF, Deschacht N, Muyldermans S. 2011. A case of convergence: Why did a simple alternative

to canonical antibodies arise in sharks and camels? PLoS Biol. 9:e100 11208. Nguyen VK, Desmyter A, Muyldermans S. 2001. Functional heavy-chain antibodies in Camelidae. Adv.

Immunol. 79:261–969. Blanc MR, Anouassi A, Abed MA, Tsikis G, Canepa S, et al. 2009. A one-step exclusion-binding proce-

dure for the purification of functional heavy-chain and mammalian-type γ-globulins from camelid sera.Biotechnol. Appl. Biochem. 54:207–12

10. Lauwereys M, Ghahroudi MA, Desmyter A, Kinne J, Holzer W, et al. 1998. Potent enzyme inhibitorsderived from dromedary heavy-chain antibodies. EMBO J. 17:3512–20

11. Daley LP, Gagliardo LF, Duffy MS, Smith MC, Appleton JA. 2005. Application of monoclonal antibodiesin functional and comparative investigations of heavy-chain immunoglobulins in New World camelids.Clin. Diagn. Lab. Immunol. 12:380–86

12. Daley LP. 2007. The significance of heavy-chain antibodies in camelid immunity. PhD diss. CornellUniv., Ithaca, NY. 151 pp.

13. Daley-Bauer LP, Purdy SR, Smith MC, Gagliardo LF, Davis WC, Appleton JA. 2012. Contributionsof conventional and heavy-chain IgG to immunity in fetal, neonatal, and adult alpacas. Clin. VaccineImmunol. 17:2007–15

14. Conrath KE, Wernery U, Muyldermans S, Nguyen VK. 2003. Emergence and evolution of functionalheavy-chain antibodies in Camelidae. Dev. Comp. Immunol. 27:87–103

15. Rabbani H, Brown WR, Butler JE, Hammarstrom L. 1997. Genetic polymorphism of bovine IgG3.Immunogenetics 46:326–31

16. Butler JE, Wertz N, Deschacht N, Kacskovics I. 2009. Porcine IgG: structure, genetics and evolution.Immunogenetics 61:209–30

17. Daley LP, Kutzler MA, Bennett BW, Smith MC, Glaser AL, Appleton JA. 2010. Effector functions ofcamelid heavy-chain antibodies in immunity to West Nile virus. Clin. Vaccine Immunol. 17:239–46

18. Chothia C, Lesk AM, Gherardi E, Tomlinson IM, Walter G, et al. 1992. Structural repertoire of thehuman VH segments. J. Mol. Biol. 227:799–817

19. Sundberg EJ, Mariuzza RA. 2002. Molecular recognition in antibody-antigen complexes. Adv. ProteinChem. 61:119–80

20. Chothia C, Novotny J, Bruccoleri R, Karplus M. 1985. Domain association in immunoglobulinmolecules: the packing of variable domains. J. Mol. Biol. 186:651–63

21. Muyldermans S, Atarhouch T, Saldanha J, Barbosa J, Hamers R. 1994. Sequence and structure of V-Hdomain from naturally-occurring camel heavy-chain immunoglobulins lacking light-chains. Protein Eng.7:1129–35

22. Vu KB, Ghahroudi MA, Wyns L, Muyldermans S. 1997. Comparison of llama V-H sequences fromconventional and heavy chain antibodies. Mol. Immunol. 34:1121–31

23. Harmsen MM, Ruuls RC, Nijman IJ, Niewold TA, Frenken LGJ, de Geus B. 2000. Llama heavy-chainV regions consist of at least four distinct subfamilies revealing novel sequence features. Mol. Immunol.37:579–90

24. Maass DR, Sepulveda J, Pernthaner A, Shoemaker CB. 2007. Alpaca (Lama pacos) as a convenient sourceof recombinant camelid heavy chain antibodies (VHHs). J. Immunol. Methods 324:13–25



Ann

u. R

ev. B

ioch

em. 2

013.

82. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by Y

ale

Uni

vers

ity -

SO

CIA

L S

CIE

NC

E L

IBR

AR

Y o

n 03

/19/

13. F

or p

erso

nal u

se o

nly.


25. Davies J, Riechmann L. 1994. ‘Camelising’ human-antibody fragments: NMR studies on VH domains.FEBS Lett. 339:285–90

26. Conrath K, Vincke C, Stijlemans B, Schymkowitz J, Decanniere K, et al. 2005. Antigen binding andsolubility effects upon veneering of a camel VHH in framework-2 to mimic a VH. J. Mol. Biol. 350:112–25

27. Jespers L, Schon O, James LC, Veprintsev D, Winter G. 2004. Crystal structure of HEL4, a soluble,refoldable human VH single domain with a germ-line scaffold. J. Mol. Biol. 337:893–903

28. This is a systematicstudy on how togenerate autonomoussingle-domain Vantibody fragmentsderived from VH-VL.

28. Barthelemy PA, Raab H, Appleton BA, Bond CJ, Wu P, et al. 2008. Comprehensive analysis ofthe factors contributing to the stability and solubility of autonomous human VH domains. J. Biol.Chem. 283:3639–54

29. Nguyen VK, Su C, Muyldermans S, van der Loo W. 2002. Heavy-chain antibodies in Camelidae; a caseof evolutionary innovation. Immunogenetics 54:39–47

30. De Genst E, Silence K, Decanniere K, Conrath K, Loris R, et al. 2006. Molecular basis for the preferentialcleft recognition by dromedary heavy-chain antibodies. Proc. Natl. Acad. Sci. USA 103:4586–91

31. Nguyen VK, Hamers R, Wyns L, Muyldermans S. 2000. Camel heavy-chain antibodies: diverse germlineVHH and specific mechanisms enlarge the antigen-binding repertoire. EMBO J. 19:921–30

32. Govaert J, Pellis M, Deschacht N, Vincke C, Conrath K, et al. 2012. Dual beneficial effect of interloopdisulfide bond for single domain antibody fragments. J. Biol. Chem. 287:1970–79

33. Stanfield RL, Dooley H, Verdino P, Flajnik MF, Wilson IA. 2007. Maturation of shark single-domain(IgNAR) antibodies: evidence for induced-fit binding. J. Mol. Biol. 367:358–72

34. This contains thefirst structure of sharkV-NAR with lysozyme,showing structuralsimilarities betweenV-NAR and VHH.

34. Stanfield RL, Dooley H, Flajnik MF, Wilson IA. 2004. Crystal structure of a shark single-domainantibody V region in complex with lysozyme. Science 305:1770–73

35. Kirchhofer A, Helma J, Schmidthals K, Frauer C, Cui S, et al. 2010. Modulation of protein propertiesin living cells using nanobodies. Nat. Struct. Mol. Biol. 17:133–38

36. Sircar A, Sanni KA, Shi J, Gray JJ. 2011. Analysis and modeling of the variable region of camelid single-domain antibodies. J. Immunol. 186:6357–67

37. This article showsthe VHH-lysozymestructure with a longprotruding VHH loopinserted in the activesite of its antigen.

37. Desmyter A, Transue TR, Ghahroudi MA, Thi MHD, Poortmans F, et al. 1996. Crystal structureof a camel single-domain V H antibody fragment in complex with lysozyme. Nat. Struct. Biol.3:803–11

38. Desmyter A, Decanniere K, Muyldermans S, Wyns L. 2001. Antigen specificity and high affinity bindingprovided by one single loop of a camel single-domain antibody. J. Biol. Chem. 276:26285–90

39. Decanniere K, Desmyter A, Lauwereys M, Ghahroudi MA, Muyldermans S, Wyns L. 1999. A single-domain antibody fragment in complex with RNase A: non-canonical loop structures and nanomolaraffinity using two CDR loops. Struct. Fold. Des. 7:361–70

40. Spinelli S, Tegoni M, Frenken L, van Vliet C, Cambillau C. 2001. Lateral recognition of a dye haptenby a llama VHH domain. J. Mol. Biol. 311:123–29

41. Spinelli S, Frenken LGJ, Hermans P, Verrips T, Brown K, et al. 2000. Camelid heavy-chain variabledomains provide efficient combining sites to haptens. Biochemistry 39:1217–22

42. Matsuda F, Honjo T. 1996. Organization of the human immunoglobulin heavy-chain locus. Adv. Im-munol. 62:1–29

43. The alpaca genomeorganization revealsthat conventionalantibodies and HCAbsgenes reside in same Hlocus.

43. Achour I, Cavelier P, Tichit M, Bouchier C, Lafaye P, Rougeon F. 2008. Tetrameric and ho-modimeric camelid IgGs originate from the same IgH locus. J. Immunol. 181:2001–9

44. Woolven BP, Frenken LGJ, van der Logt P, Nicholls PJ. 1999. The structure of the llama heavy chainconstant genes reveals a mechanism for heavy-chain antibody formation. Immunogenetics 50:98–101

45. Nguyen VK, Hamers R, Wyns L, Muyldermans S. 1999. Loss of splice consensus signal is responsiblefor the removal of the entire CH1 domain of the functional camel IGG2A heavy-chain antibodies. Mol.Immunol. 36:515–24

46. Nguyen VK, Zou XG, Lauwereys M, Brys L, Bruggemann M, Muyldermans S. 2003. Heavy-chain onlyantibodies derived from dromedary are secreted and displayed by mouse B cells. Immunology 109:93–101

47. Zou XG, Smith JA, Nguyen VK, Ren LM, Luyten K, et al. 2005. Expression of a dromedary heavychain-only antibody and B cell development in the mouse. J. Immunol. 175:3769–79

48. Janssens R, Dekker S, Hendriks RW, Panayotou G, van Remoortere A, et al. 2006. Generation ofheavy-chain-only antibodies in mice. Proc. Natl. Acad. Sci. USA 103:15130–35

17.18 Muyldermans


Ann

u. R

ev. B

ioch

em. 2

013.

82. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by Y

ale

Uni

vers

ity -

SO

CIA

L S

CIE

NC

E L

IBR

AR

Y o

n 03

/19/

13. F

or p

erso

nal u

se o

nly.


49. Nguyen VK, Muyldermans S, Hamers R. 1998. The specific variable domain of camel heavy-chainantibodies is encoded in the germline. J. Mol. Biol. 275:413–18

50. Ben Abderrazek R, Vincke C, Hmila I, Saerens D, Abidi N, et al. 2011. Development of Cys38 knock-outand humanized version of NbAahII10 nanobody with improved neutralization of AahII scorpion toxin.Protein Eng. Des. Sel. 24:727–35

51. Al-Lazikani B, Lesk AM, Chothia C. 1997. Standard conformations for the canonical structures ofimmunoglobulins. J. Mol. Biol. 273:927–48

52. Decanniere K, Muyldermans S, Wyns L. 2000. Canonical antigen-binding loop structures in im-munoglobulins: More structures, more canonical classes? J. Mol. Biol. 300:83–91

53. De Genst E, Silence K, Ghahroudi MA, Decanniere K, Loris R, et al. 2005. Strong in vivo maturationcompensates for structurally restricted H3 loops in antibody repertoires. J. Biol. Chem. 280:14114–21

54. Deschacht N, De Groeve K, Vincke C, Raes G, De Baetselier P, Muyldermans S. 2010. A novel promis-cuous class of camelid single-domain antibody contributes to the antigen-binding repertoire. J. Immunol.184:5696–704

55. Bankovich AJ, Raunser S, Juo ZS, Walz T, Davis MM, Garcia KC. 2007. Structural insight into pre-Bcell receptor function. Science 316:291–94

56. Feige MJ, Groscurth S, Marcinowski M, Shimizu Y, Kessler H, et al. 2009. An unfolded CH1 domaincontrols the assembly and secretion of IgG antibodies. Mol. Cell 34:569–79

57. Martensson IL, Almqvist N, Grimsholm O, Bernardi AI. 2010. The pre-B cell receptor checkpoint.FEBS Lett. 584:2572–79

58. Holliger P, Hudson PJ. 2005. Engineered antibody fragments and the rise of single domains. Nat.Biotechnol. 23:1126–36

59. Koch-Nolte F, Reyelt J, Schoessow B, Schwarz N, Scheuplein F, et al. 2007. Single domain antibodiesfrom llama effectively and specifically block T cell ecto-ADP-ribosyltransferase ART2.2 in vivo. FASEBJ. 21:3490–98

60. Monegal A, Ami D, Martinelli C, Huang H, Aliprandi M, et al. 2009. Immunological applications ofsingle-domain llama recombinant antibodies isolated from a naive library. Protein Eng. Des. Sel. 22:273–80

61. Goldman ER, Anderson GP, Liu JL, Delehanty JB, Sherwood LJ, et al. 2006. Facile generation of heat-stable antiviral and antitoxin single domain antibodies from a semisynthetic llama library. Anal. Chem.78:8245–55

62. Pellis M, Pardon E, Zolghadr K, Rothbauer U, Vincke C, et al. 2012. A bacterial-two-hybrid selectionsystem for one-step isolation of intracellularly functional Nanobodies. Arch. Biochem. Biophys. 526:114–23

63. Zolghadr K, Mortusewicz O, Rothbauer U, Kleinhans R, Goehler H, et al. 2008. A fluorescent two-hybrid assay for direct visualization of protein interactions in living cells. Mol. Cell. Proteomics 7:2279–87

64. Yau KYF, Groves MAT, Li SH, Sheedy C, Lee H, et al. 2003. Selection of hapten-specific single-domainantibodies from a non-immunized llama ribosome display library. J. Immunol. Methods 281:161–75

65. Hoogenboom HR. 2005. Selecting and screening recombinant antibody libraries. Nat. Biotechnol.23:1105–16

66. Baral TN, Murad Y, Thanh-Dung N, Iqbal U, Zhang J. 2011. Isolation of functional single domainantibody by whole cell immunization: implications for cancer treatment. J. Immunol. Methods 371:70–80

67. Deckers N, Saerens D, Kanobana K, Conrath K, Victor B, et al. 2009. Nanobodies, a promising tool forspecies-specific diagnosis of Taenia solium cysticercosis. Int. J. Parasitol. 39:625–33

68. Yau KYF, Dubuc G, Li SH, Hirama T, MacKenzie CR, et al. 2005. Affinity maturation of a VHH bymutational hotspot randomization. J. Immunol. Methods 297:213–24

69. Koide A, Tereshko V, Uysal S, Margalef K, Kossiakoff AA, Koide S. 2007. Exploring the capacity ofminimalist protein interfaces: interface energetics and affinity maturation to picomolar KD of a single-domain antibody with a flat paratope. J. Mol. Biol. 373:941–53

70. De Genst E, Areskoug D, Decanniere K, Muyldermans S, Andersson K. 2002. Kinetic and affinity predic-tions of a protein-protein interaction using multivariate experimental design. J. Biol. Chem. 277:29897–907

71. Arbabi-Ghahroudi M, Tanha J, MacKenzie R. 2005. Prokaryotic expression of antibodies. Cancer Metas-tasis Rev. 24:501–19



Ann

u. R

ev. B

ioch

em. 2

013.

82. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by Y

ale

Uni

vers

ity -

SO

CIA

L S

CIE

NC

E L

IBR

AR

Y o

n 03

/19/

13. F

or p

erso

nal u

se o

nly.


72. Frenken LGJ, van der Linden RHJ, Hermans P, Bos JW, Ruuls RC, et al. 2000. Isolation of anti-gen specific llama V-HH antibody fragments and their high level secretion by Saccharomyces cerevisiae.J. Biotechnol. 78:11–21

73. Frenken LGJ, Hessing JGM, Van den Hondel C, Verrips CT. 1998. Recent advances in the large-scaleproduction of antibody fragments using lower eukaryotic microorganisms. Res. Immunol. 149:589–99

74. Ismaili A, Jalali-Javaran M, Rasaee MJ, Rahbarizadeh F, Forouzandeh-Moghadam M, Memari HR. 2007.Production and characterization of anti-(mucin MUCI) single-domain antibody in tobacco (Nicotianatabacum cultivar Xanthi). Biotechnol. Appl. Biochem. 47:11–19

75. Graille M, Stura EA, Corper AL, Sutton BJ, Taussig MJ, et al. 2000. Crystal structure of a Staphylococcusaureus protein A domain complexed with the Fab fragment of a human IgM antibody: structural basisfor recognition of B-cell receptors and superantigen activity. Proc. Natl. Acad. Sci. USA 97:5399–404

76. Ghahroudi MA, Desmyter A, Wyns L, Hamers R, Muyldermans S. 1997. Selection and identificationof single domain antibody fragments from camel heavy-chain antibodies. FEBS Lett. 414:521–26

77. Dumoulin M, Conrath K, Van Meirhaeghe A, Meersman F, Heremans K, et al. 2002. Single-domainantibody fragments with high conformational stability. Protein Sci. 11:500–15

78. van der Linden RHJ, Frenken LGJ, de Geus B, Harmsen MM, Ruuls RC, et al. 1999. Comparisonof physical chemical properties of llama VHH antibody fragments and mouse monoclonal antibodies.Biochim. Biophys. Acta 1431:37–46

79. Hagihara Y, Mine S, Uegaki K. 2007. Stabilization of an immunoglobulin fold domain by an engineereddisulfide bond at the buried hydrophobic region. J. Biol. Chem. 282:36489–95

80. Saerens D, Conrath K, Govaert J, Muyldermans S. 2008. Disulfide bond introduction for general stabi-lization of immunoglobulin heavy-chain variable domains. J. Mol. Biol. 377:478–88

81. Hussack G, Hirama T, Ding W, MacKenzie R, Tanha J. 2011. Engineered single-domain antibodieswith high protease resistance and thermal stability. PLoS ONE 6:e28218

82. Harmsen MM, van Solt CB, van Zijderveld-van Bemmel AM, Niewold TA, van Zijderveld FG. 2006.Selection and optimization of proteolytically stable llama single-domain antibody fragments for oralimmunotherapy. Appl. Microbiol. Biotechnol. 72:544–51

83. Dolk E, van der Vaart M, Hulsik DL, Vriend G, de Haard H, et al. 2005. Isolation of llama antibodyfragments for prevention of dandruff by phage display in shampoo. Appl. Environ. Microbiol. 71:442–50

84. Christ D, Famm K, Winter G. 2007. Repertoires of aggregation-resistant human antibody domains.Protein Eng. Des. Sel. 20:413–16

85. Jespers L, Schon O, Famm K, Winter G. 2004. Aggregation-resistant domain antibodies selected onphage by heat denaturation. Nat. Biotechnol. 22:1161–65

86. Saerens D, Pellis M, Loris R, Pardon E, Dumoulin M, et al. 2005. Identification of a universal VHHframework to graft non-canonical antigen-binding loops of camel single-domain antibodies. J. Mol. Biol.352:597–607

87. Dudgeon K, Famm K, Christ D. 2009. Sequence determinants of protein aggregation in human VHdomains. Protein Eng. Des. Sel. 22:217–20

88. Bond CJ, Marsters JC, Sidhu SS. 2003. Contributions of CDR3 to VHH domain stability and the designof monobody scaffolds for naive antibody libraries. J. Mol. Biol. 332:643–55

89. Cortez-Retamozo V, Backmann N, Senter PD, Wernery U, De Baetselier P, et al. 2004. Efficient cancertherapy with a nanobody-based conjugate. Cancer Res. 64:2853–57

90. Baral TN, Magez S, Stijlemans B, Conrath K, Vanhollebeke B, et al. 2006. Experimental therapy ofAfrican trypanosomiasis with a nanobody-conjugated human trypanolytic factor. Nat. Med. 12:580–84

91. Coppieters K, Dreier T, Silence K, de Haard H, Lauwereys M, et al. 2006. Formatted anti-tumor necrosisfactor α VHH proteins derived from camelids show superior potency and targeting to inflamed jointsin a murine model of collagen-induced arthritis. Arthritis Rheum. 54:1856–66

92. Vincke C, Loris R, Saerens D, Martinez-Rodriguez S, Muyldermans S, Conrath K. 2009. Generalstrategy to humanize a camelid single-domain antibody and identification of a universal humanizednanobody scaffold. J. Biol. Chem. 284:3273–84

93. Transue TR, De Genst E, Ghahroudi MA, Wyns L, Muyldermans S. 1998. Camel single-domain anti-body inhibits enzyme by mimicking carbohydrate substrate. Proteins 32:515–22

17.20 Muyldermans


Ann

u. R

ev. B

ioch

em. 2

013.

82. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by Y

ale

Uni

vers

ity -

SO

CIA

L S

CIE

NC

E L

IBR

AR

Y o

n 03

/19/

13. F

or p

erso

nal u

se o

nly.


94. Hendrickx MLV, De Winter A, Buelens K, Compernolle G, Hassanzadeh-Ghassabeh G, et al. 2011.TAFIa inhibiting nanobodies as profibrinolytic tools and discovery of a new TAFIa conformation.J. Thromb. Haemost. 9:2268–77

95. Conrath KE, Lauwereys M, Galleni M, Matagne A, Frere JM, et al. 2001. β-Lactamase inhibitorsderived from single-domain antibody fragments elicited in the Camelidae. Antimicrob. Agents Chemother.45:2807–12

96. Oyen D, Srinivasan V, Steyaert J, Barlow JN. 2011. Constraining enzyme conformational change by anantibody leads to hyperbolic inhibition. J. Mol. Biol. 407:138–48

97. Barlow JN, Conrath K, Steyaert J. 2009. Substrate-dependent modulation of enzyme activity by allostericeffector antibodies. Biochim. Biophys. Acta 1794:1259–68

98. Saerens D, Kinne J, Bosmans E, Wernery U, Muyldermans S, Conrath K. 2004. Single domain antibodiesderived from dromedary lymph node and peripheral blood lymphocytes sensing conformational variantsof prostate-specific antigen. J. Biol. Chem. 279:51965–72

99. Saerens D, Ghassabeh GH, Muyldermans S. 2008. Single-domain antibodies as building blocks for noveltherapeutics. Curr. Opin. Pharmacol 8:600–8

100. Emmerson CD, van der Vlist EJ, Braam MR, Vanlandschoot P, Merchiers P, et al. 2011. Enhancementof polymeric immunoglobulin receptor transcytosis by biparatopic VHH. PLoS ONE 6:e26299

101. Conrath KE, Lauwereys M, Wyns L, Muyldermans S. 2001. Camel single-domain antibodies as modularbuilding units in bispecific and bivalent antibody constructs. J. Biol. Chem. 276:7346–50

102. Chames P, Baty D. 2009. Bispecific antibodies for cancer therapy. Curr. Opin. Drug Discov. Dev. 12:276–83

103. Zhang JB, Tanha J, Hirama T, Khieu NH, To R, et al. 2004. Pentamerization of single-domain antibodiesfrom phage libraries: a novel strategy for the rapid generation of high-avidity antibody reagents. J. Mol.Biol. 335:49–56

104. Bell A, Wang ZJ, Arbabi-Ghahroudi M, Chang TA, Durocher Y, et al. 2010. Differential tumor-targetingabilities of three single-domain antibody formats. Cancer Lett. 289:81–90

105. Hmila I, Abdallah BA-B, Saerens D, Benlasfar Z, Conrath K, et al. 2008. VHH, bivalent domainsand chimeric heavy chain-only antibodies with high neutralizing efficacy for scorpion toxin AahI’. Mol.Immunol. 45:3847–56

106. Harmsen M, De Haard H. 2007. Properties, production, and applications of camelid single-domainantibody fragments. Appl. Microb. Biotechnol. 77:13–22

107. This article is acomprehensiveoverview ofnanobody-basedapplications.

107. Wesolowski J, Alzogaray V, Reyelt J, Unger M, Juarez K, et al. 2009. Single domain antibod-ies: promising experimental and therapeutic tools in infection and immunity. Med. Microbiol.Immunol. 198:157–74

108. de Marco A. 2011. Biotechnological applications of recombinant single-domain antibody fragments.Microb. Cell Factories 10:44

109. Taussig MJ, Stoevesandt O, Borrebaeck CAK, Bradbury AR, Cahill D, et al. 2007. ProteomeBinders:planning a European resource of affinity reagents for analysis of the human proteome. Nat. Methods4:13–17

110. This was the firstpaper usingfluorescently labelednanobodies to traceantigen in living cells.

110. Rothbauer U, Zolghadr K, Tillib S, Nowak D, Schermelleh L, et al. 2006. Targeting and tracingantigens in live cells with fluorescent nanobodies. Nat. Methods 3:887–89

111. Olichon A, Surrey T. 2007. Selection of genetically encoded fluorescent single domain antibodies engi-neered for efficient expression in Escherichia coli. J. Biol. Chem. 282:36314–20

112. Ries J, Kaplan C, Platonova E, Eghlidi H, Ewers H. 2012. A simple, versatile method for GFP-basedsuper-resolution microspcopy via nanobodies. Nat. Methods 9:582–84

113. Romer T, Leonhardt H, Rothbauer U. 2011. Engineering antibodies and proteins for molecular in vivoimaging. Curr. Opin. Biotechnol. 22:882–87

114. Schmidthals K, Helma J, Zolghadr K, Rothbauer U, Leonhardt H. 2010. Novel antibody derivatives forproteome and high-content analysis. Anal. Bioanal. Chem. 397:3203–8

115. Rothbauer U, Zolghadr K, Muyldermans S, Schepers A, Cardoso MC, Leonhardt H. 2008. A versatilenanotrap for biochemical and functional studies with fluorescent fusion proteins. Mol. Cell. Proteomics7:282–89



Ann

u. R

ev. B

ioch

em. 2

013.

82. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by Y

ale

Uni

vers

ity -

SO

CIA

L S

CIE

NC

E L

IBR

AR

Y o

n 03

/19/

13. F

or p

erso

nal u

se o

nly.


116. Tanaka T, Sewell H, Waters S, Phillips SEV, Rabbitts TH. 2011. Single domain intracellular antibodiesfrom diverse libraries: emphasizing dual functions of LMO2 protein interactions using a single VHdomain. J. Biol. Chem. 286:3707–16

117. Perez-Martinez D, Tanaka T, Rabbitts TH. 2010. Intracellular antibodies and cancer: new technologiesoffer therapeutic opportunities. BioEssays 32:589–98118. This describes the

method to turnnanobodies intoprotein-degradingmachines to knock outspecifically targetedproteins.

118. Caussinus E, Kanca O, Affolter M. 2012. Fluorescent fusion protein knockout mediated byanti-GFP nanobody. Nat. Struct. Mol. Biol. 19:117–21

119. This reviewsprogress in usingtarget-specific reagents,such as nanobodies, ascrystallizingchaperones.

119. Koide S. 2009. Engineering of recombinant crystallization chaperones. Curr. Opin. Struct. Biol.19:449–57

120. Baranova E, Fronzes R, Garcia-Pino A, Van Gerven N, Papapostolou D, et al. 2012. SbsB structure andlattice reconstruction unveil Ca2+ triggered S-layer assembly. Nature 487:119–22

121. Rasmussen SGF, Choi H-J, Fung JJ, Pardon E, Casarosa P, et al. 2011. Structure of a nanobody-stabilizedactive state of the β2 adrenoceptor. Nature 469:175–80

122. Steyaert J, Kobilka BK. 2011. Nanobody stabilization of G protein–coupled receptor conformationalstates. Curr. Opin. Struct. Biol. 21:567–72

123. Korotkov KV, Pardon E, Steyaert J, Hol WGJ. 2009. Crystal structure of the N-terminal domain of thesecretin GspD from ETEC determined with the assistance of a nanobody. Structure 17:255–65

124. Saerens D, Frederix F, Reekmans G, Conrath K, Jans K, et al. 2005. Engineering camel single-domain an-tibodies and immobilization chemistry for human prostate-specific antigen sensing. Anal. Chem. 77:7547–55

125. Van de Broek B, Devoogdt N, D’Hollander A, Gijs H-L, Jans K, et al. 2011. Specific cell targeting withnanobody conjugated branched gold nanoparticles for photothermal therapy. ACS Nano 5:4319–28

126. Pollithy A, Romer T, Lang C, Muller FD, Helma J, et al. 2011. Magnetosome expression of functionalcamelid antibody fragments (nanobodies) in Magnetospirillum gryphiswaldense. Appl. Environ. Microbiol.77:6165–71

127. Klooster R, Maassen BTH, Stam JC, Hermans PW, ten Haaft MR, et al. 2007. Improved anti-IgG andHSA affinity ligands: clinical application of VHH antibody technology. J. Immunol. Methods 324:1–12

128. De Genst EJ, Guilliams T, Wellens J, O’Day EM, Waudby CA, et al. 2010. Structure and properties ofa complex of α-synuclein and a single-domain camelid antibody. J. Mol. Biol. 402:326–43

129. Vaneycken I, D’Hhuyvetter M, Hernot S, De Vos J, Xavier C, et al. 2011. Immuno-imaging usingnanobodies. Curr. Opin. Biotechnol. 22:877–81

130. Vosjan MJWD, Perk LR, Roovers RC, Visser GWM, Stigter-van Walsum M, et al. 2011. Facile labellingof an anti-epidermal growth factor receptor nanobody with 68Ga via a novel bifunctional desferal chelatefor immuno-PET. Eur. J. Nuclear Med. Mol. Imaging 38:753–63

131. Vaneycken I, Devoogdt N, Van Gassen N, Vincke C, Xavier C, et al. 2011. Preclinical screening ofanti-HER2 nanobodies for molecular imaging of breast cancer. FASEB J. 25:2433–46

132. This illustrateshow nanobodies can beused for noninvasive invivo imaging.

132. Broisat A, Hernot S, Toczek J, De Vos J, Riou LM, et al. 2012. Nanobodies targetingmouse/human VCAM1 for the nuclear imaging of atherosclerotic lesions. Circ. Res. 110:927–37

133. Hmila I, Saerens D, Ben Abderrazek R, Vincke C, Abidi N, et al. 2010. A bispecific nanobody to providefull protection against lethal scorpion envenoming. FASEB J. 24:3479–89

134. Hussack G, Arbabi-Ghahroudi M, van Faassen H, Songer JG, Ng KKS, et al. 2011. Neutralization ofClostridium difficile toxin A with single-domain antibodies targeting the cell receptor binding domain.J. Biol. Chem. 286:8961–76

135. Stewart CS, MacKenzie CR, Hall JC. 2007. Isolation, characterization and pentamerization of α-cobrotoxin specific single-domain antibodies from a naıve phage display library: preliminary findingsfor antivenom development. Toxicon 49:699–709

136. Cook DAN, Samarasekara CL, Wagstaff SC, Kinne J, Wernery U, Harrison RA. 2010. Analysis ofcamelid IgG for antivenom development: Immunoreactivity and preclinical neutralisation of venom-induced pathology by IgG subclasses, and the effect of heat treatment. Toxicon 56:596–603

137. Harrison RA, Hasson SS, Harmsen M, Laing GD, Conrath K, Theakston RDG. 2006. Neutralisationof venom-induced haemorrhage by IgG from camels and llamas immunised with viper venom and alsoby endogenous, non-IgG components in camelid sera. Toxicon 47:364–68

17.22 Muyldermans


Ann

u. R

ev. B

ioch

em. 2

013.

82. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by Y

ale

Uni

vers

ity -

SO

CIA

L S

CIE

NC

E L

IBR

AR

Y o

n 03

/19/

13. F

or p

erso

nal u

se o

nly.


138. Pant N, Marcotte H, Hermans P, Bezemer S, Frenken L, et al. 2011. Lactobacilli producing bispecificllama-derived anti-rotavirus proteins in vivo for rotavirus-induced diarrhea. Future Microbiol. 6:583–93

139. Kruger C, Hultberg A, Marcotte H, Hermans P, Bezemer S, et al. 2006. Therapeutic effect of llamaderived VHH fragments against Streptococcus mutans on the development of dental caries. Appl. Microbiol.Biotechnol. 72:732–7

140. Harmsen MM, Fijten HPD, Engel B, Dekker A, Eble PL. 2009. Passive immunization with llamasingle-domain antibody fragments reduces foot-and-mouth disease transmission between pigs. Vaccine27:1904–11

141. Strokappe N, Szynol A, Aasa-Chapman M, Gorlani A, Quigley AF, et al. 2012. Llama antibody fragmentsrecognizing various epitopes of the CD4bs neutralize a broad range of HIV-1 subtypes A, B and C. PLoSONE 7:e33298

142. Koh WWL, Steffensen S, Gonzalez-Pajuelo M, Hoorelbeke B, Gorlani A, et al. 2010. Generation ofa family-specific phage library of llama single chain antibody fragments that neutralize HIV-1. J. Biol.Chem. 285:19116–24

143. Hinz A, Hulsik DL, Forsman A, Koh WWL, Belrhali H, et al. 2010. Crystal structure of the neutralizingllama V-HH D7 and its mode of HIV-1 gp120 interaction. PLoS ONE 5:e10482

144. De Haard HJW, Bezemer S, Ledeboer AM, Muller WH, Boender PJ, et al. 2005. Llama antibodiesagainst a lactococcal protein located at the tip of the phage tail prevent phage infection. J. Bacteriol.187:4531–41

145. Ibanez LI, De Filette M, Hultberg A, Verrips T, Temperton N, et al. 2011. Nanobodies with in vitroneutralizing activity protect mice against H5N1 influenza virus infection. J. Infect. Dis. 203:1063–72

146. Stijlemans B, Conrath K, Cortez-Retamozo V, Van Xong H, Wyns L, et al. 2004. Efficient targetingof conserved cryptic epitopes of infectious agents by single domain antibodies: African trypanosomes asparadigm. J. Biol. Chem. 279:1256–61

147. Holz J-B. 2012. The TITAN trial—assessing the efficacy and safety of an anti-von Willebrand fac-tor nanobody in patients with acquired thrombotic thrombocytopenic purpura. Transfus. Apheresis Sci.46:343–46

148. Kontermann RE. 2011. Strategies for extended serum half-life of protein therapeutics. Curr. Opin.Biotechnol. 22:868–76

149. Li T, Bourgeois JP, Celli S, Glacial F, Le Sourd AM, et al. 2012. Cell-penetrating anti-GFAP VHH andcorresponding fluorescent fusion protein VHH-GFP spontaneously cross the blood-brain barrier andspecifically recognize astrocytes: application to brain imaging. FASEB J. 26:3969–79

150. Perruchini C, Pecorari F, Bourgeois J-P, Duyckaerts C, Rougeon F, Lafaye P. 2009. Llama VHHantibody fragments against GFAP: better diffusion in fixed tissues than classical monoclonal antibodies.Acta Neuropathol. 118:685–95

151. Iqbal U, Trojahn U, Albaghdadi H, Zhang J, O’Connor-McCourt M, et al. 2010. Kinetic analysis ofnovel mono- and multivalent VHH-fragments and their application for molecular imaging of braintumours. Br. J. Pharmacol. 160:1016–28



Ann

u. R

ev. B

ioch

em. 2

013.

82. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by Y

ale

Uni

vers

ity -

SO

CIA

L S

CIE

NC

E L

IBR

AR

Y o

n 03

/19/

13. F

or p

erso

nal u

se o

nly.

Documents

Nanobodies: Natural Single-Domain Antibodies€¦ · Nanobodies: Natural Single-Domain Antibodies Serge Muyldermans Research Group, Cellular and Molecular Immunology, Vrije Universiteit