Recombinant chimeric lectins consisting of mannose-binding lectin and L-ficolin are potent inhibitors of influenza A virus compared with mannose-binding lectin.

Abstract

MBL structurally contains a type II-like collagenous domain and a carbohydrate recognition domain (CRD). We have recently generated three novel recombinant chimeric lectins (RCL), in which varying length of collagenous domain of mannose-binding lectin (MBL) is replaced with that of L-ficolin (L-FCN). CRD of MBL is used for target recognition because it has a broad spectrum in pathogen recognition compared with L-FCN. Results of our study demonstrate that these RCLs are potent inhibitors of influenza A virus (IAV). RCLs, against IAV, show dose-dependent activation of the lectin complement pathway, which is significantly higher than that of recombinant human MBL (rMBL). This activity is observed even without MBL-associated serine proteases (MASPs, provided by MBL deficient mouse sera), which have been thought to mediate complement activation. These observations suggest that RCLs are more efficient in associating with MASP-2, which predominantly mediates the activity. Yet, additional serum further increases the activity while RCL-mediated coagulation-like enzyme activities are diminished compared with rMBL, suggesting reduced association with MASP-1, which has been shown to mediate coagulation-like activity. These data suggest that RCLs may interfere less with host coagulation, which is advantageous to be a therapeutic drug. Importantly, these RCLs have surpassed rMBL for anti-viral activities, such as viral aggregation, reduction of viral hemagglutination (HA) and inhibition of virus-mediated HA and neuraminidase (NA) activities. These results are encouraging that novel RCLs could be used as anti-IAV agents with less side effect and that RCLs would be suitable candidates in developing a new anti-IAV therapy.

Full text

Page 1

Recombinant chimeric lectins consisting of mannose-binding lectin and L-ficolin
are potent inhibitors of influenza A virus compared with mannose-binding lectin
Wei-Chuan Changa,1, Kevan L. Hartshornb,1, Mitchell R. Whiteb, Patience Moyoa, Ian C. Michelowa,
Henry Kozielc, Bernard T. Kinanea, Emmett V. Schmidtd, Teizo Fujitae, Kazue Takahashia,*
aProgram of Developmental Immunology, Department of Pediatrics, Massachusetts General Hospital, Harvard Medical School, Boston, MA 02114, United States
bDepartment of Medicine, Boston University School of Medicine, Boston, MA 02118, United States
cDivisionofPulmonary,CriticalCare,andSleepMedicine,DepartmentofMedicine,BethIsraelDeaconessMedicalCenterandHarvardMedicalSchool,Boston,MA02115,UnitedStates
dCancer Center, Massachusetts General Hospital, Harvard Medical School, Boston, MA 02114, United States
eDepartment of Immunology, Fukushima Medical University, Fukushima 960-1295, Japan
1. Introduction
IAV is an RNA virus whose surface is enveloped with
glycoproteins containing neuraminidase (NA) and hemagglutinin
(HA), which have glycosylation sites [1]. IAV infection is a common
infection that could result in fatal complications, even in
individuals who are appeared to be healthy [2,3]. Mortality and
hospitalization are estimated to exceed annually 30,000 and
200,000, respectively in the United States alone [2]. Prevention is
currently relied upon immunization, however vaccines are less
effective in elderly and are not approved by the FDA for infants
younger than 6 months of age [3,4]. Resistance to antiviral agents
has developed in seasonal and pandemic IAV strains [2,5]. Thus,
there is a need for new effective anti-IAV therapeutics.
The first line of host defense is the innate immunity, which
recognizespathogen-associated
through pattern recognition receptors and soluble molecules that
include lectins [6]. One such lectin is MBL, which is primarily
synthesized in the liver and circulates in the blood [7–9]. MBL
belongs to the collectin family that is structurally characterized by
consisting of a type II-like collagenous domain at N-terminus,
followed by a neck region and a carbohydrate recognition domain
(CRD) at C-terminus [10]. The collectin family includes lung
surfactant protein (SP)-A and SP-D [11]. These surfactant proteins
have anti-viral functions [12–17] and mice lacking SP-A or SP-D
have increased susceptibility to AIV infection [14,18].
In early 1990, MBL was identified as a b-inhibitor, which had
been discovered as an IAV inactivating serum factor in the 1940s
molecularpatterns (PAMPs)
Biochemical Pharmacology 81 (2011) 388–395
A R T I C L EI N F O
Article history:
Received 6 August 2010
Accepted 19 October 2010
Keywords:
Innate immunity
Mannose-binding lectin
Ficolin
Influenza A virus
Complement
Coagulation
A B S T R A C T
MBL structurally contains a type II-like collagenous domain and a carbohydrate recognition domain
(CRD). We have recently generated three novel recombinant chimeric lectins (RCL), in which varying
length of collagenous domain of mannose-binding lectin (MBL) is replaced with that of L-ficolin (L-FCN).
CRD of MBL is used for target recognition because it has a broad spectrum in pathogen recognition
compared with L-FCN. Results of our study demonstrate that these RCLs are potent inhibitors of
influenza A virus (IAV). RCLs, against IAV, show dose-dependent activation of the lectin complement
pathway, which is significantly higher than that of recombinant human MBL (rMBL). This activity is
observedeven withoutMBL-associated serineproteases (MASPs,providedbyMBLdeficientmouse sera),
which have been thought to mediate complement activation. These observations suggest that RCLs are
more efficient in associating with MASP-2, which predominantly mediates the activity. Yet, additional
serum further increases the activity while RCL-mediated coagulation-like enzyme activities are
diminished compared with rMBL, suggesting reduced association with MASP-1, which has been shown
to mediate coagulation-like activity. These data suggest that RCLs may interfere less with host
coagulation, which is advantageous to be a therapeutic drug. Importantly, these RCLs have surpassed
rMBL for anti-viral activities, such as viral aggregation, reduction of viral hemagglutination (HA) and
inhibition of virus-mediated HA and neuraminidase (NA) activities. These results are encouraging that
novel RCLs could be used as anti-IAV agents with less side effect and that RCLs would be suitable
candidates in developing a new anti-IAV therapy.
? 2010 Elsevier Inc. All rights reserved.
* Corresponding author at: 55 Fruit Street, GRJ1402, Boston, MA 02114, United
States. Tel.: +1 617 726 1394; fax: +1 617 724 3248.
E-mail address: ktakahashi1@partners.org (K. Takahashi).
1Co-first authors.
Contents lists available at ScienceDirect
Biochemical Pharmacology
journal homepage: www.elsevier.com/locate/biochempharm
0006-2952/$ – see front matter ? 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.bcp.2010.10.012

Page 2

[19]. Since then, many studies have described MBL’s anti-IAV
functions, including inhibition of viral hemagglutination, inhibi-
tion of HA and NA and viral neutralization [20–23]. MBL activates
complement via the lectin complement pathway, which is a key
biologic function along with other complement pathways, such as
the classical and the alternative complement pathway. The
classical complement pathway is mediated by C1rs proteases,
which are replaced by MASP-1, MASP-2 and/or MASP-3, in the
lectin complement pathway [24]. Both pathways cleave C4 and C2
to generate the C3 convertase, C4bC2a [25,26].
MBL–MASP complex also initiates coagulation via thrombin-
like activity [27,28]. Coagulation is a primitive yet effective host
defense mechanism. For example, tachylectins in horseshoe crab
hemolymph provide immune protection by clotting lipopolysac-
charide and b-glucan, pattern recognition molecules of pathogens
(PAMPs of Gram negative bacteria and fungus, respectively) [29].
The collectin family also includes L-FCN and H-FCN, which
are also circulating serum pattern recognition molecules of
innate immune system [30]. Like MBL, both FCNs contain the
collagenous domain while CRD is replaced with fibrinogen-like
domain, which preferentially recognizes acetylated molecules
and sialic acid [31,32]. In contrast, MBL’s target recognition is
broad, including mannose, which is widely expressed on many
pathogens [33]. It has been shown that other chimeric lectins
consisting of MBL-CRD and the collagenous domain of SP-D gain
anti-IAV activities, such as viral aggregation and inhibition of
HA, NA and viral infectivity [22,34].
We have previously generated three RCLs consisting of L-FCN
and MBL, in which various lengths of the collagenous domain were
replaced with that of L-FCN [23]. Previous characterization study
has demonstrated that these RCLs are either comparable to or
surpassed rMBL for several biologic activities, including their
binding to Nipah, Hendra and Ebola viruses [23]. Here, we further
characterized biologic activities of these recombinant lectins
against IAV using in vitro system and will discuss our findings.
2. Materials and methods
2.1. Recombinant chimeric lectins
Chimeric lectins were produced as previously described [23]. In
this study, these lectins are named RCL1, RCL2 and RCL3,
corresponding to L-FCN/MBL126,
MBL64, respectively in the previous publication. All RCLs have
MBL-CRDwhileMBL-collagenousdomainwasreplacedwith126,76
or 64 amino acids of L-FCN’s collagenous domain, resulting in total
amino acid length of 251, 255 or 254 for RCL1, RCL2 or RCL3,
respectively. Thus, overall amino acid length is similar while RCL1
has the longest L-FCN collagenous domain followed by RCL2 and
then RCL3. The junction of two proteins in RCL2 is located at the
middle of a putative MASP-binding domain.
L-FCN/MBL76andL-FCN/
2.2. Virus preparations
IAV (A/Phillipines/82(H3N2)) was prepared as previously
described [35]. Briefly, IAV was grown in the chorioallantoic fluid
of chicken eggs and purified on a discontinuous sucrose gradient
(Sigma–Aldrich, St. Louis, MO). Virus stocks were dialyzed against
PBS (Sigma–Aldrich, St. Louis, MO) and aliquots were stored at
?80 8C. HA titers were determined by titration with human type O,
Rh?red blood cells (RBCs) in PBS.
2.3. MBL binding assay
This assay was performed using previously described methods
with a minor modification [36]. IAV concentration was arbitrary
definedas1000 U/ml,whichwasdeterminedtobeoptimalformany
invitrostudies based ondoseresponseexperiments. Briefly, 96well
plates were coated with mannan (Sigma–Aldrich, St. Louis, MO) or
IAV and then blocked. Following wash, the wells were incubated
with indicated concentrations of recombinant lectins. After wash,
boundMBLwasdetectedbymouseanti-hMBLmonoclonalAb(2A9,
agiftfromDr.GregoryStahl)[37],followedbyalkaline-phosphatase
conjugated anti-mouse Ab (Promega, Madison, WI) and pNTP
substrate (Sigma–Aldrich, St. Louis, MO). Reaction was read at
415 nm using SpectraMax M5 (Molecular Devices, Sunnyvale, CA)
and expressed as OD 415 nm reading. Assays were performed in
triplicates and were repeated at least twice.
2.4. Mouse sera
MBL null mice were previously generated and fully backcrossed
onto C57Black/6J [36,38]. Sera were collected and stored at ?80 8C
prior to the study. All animal experiments were performed under a
protocol approved by the Subcommittee on Research Animal Care
at Massachusetts General Hospital, Boston, MA.
2.5. Assays of the lectin complement activity
The lectin pathway assay was performed with a minor
modification of previously described method [36]. Briefly, 96 well
plates were coated with mannan or IAV as above. After wash and
block, the wells were incubated with various concentrations of
lectins with orwithout1% MBLnull sera (MASP source) diluted ina
binding buffer, 10 mM Tris, pH 7.8, 10 mM CaCl2, 1 M NaCl (all
chemicals were purchased from Sigma–Aldrich, St. Louis, MO).
After wash, the wells were incubated with human C4 and
incubated at 37 8C. After wash, the wells were incubated with
rabbit anti-hC4c Ab (Dako, Carpinteria, CA) followed by biotin-
conjugated anti-rabbit Ab, alkaline phosphatase-conjugated bio-
tin–avidin (ABC-AP system, Vector Labs, Burlingame, CA) and then
with pNTP (Sigma–Aldrich, St. Louis, MO). The plates were read at
415 nm. Binding activity was expressed as OD 415 nm reading.
Pooled human serum with known MBL concentration and C4
activity, which was arbitrarily defined as 1000 U/ml (State Serum
Institute, Denmark), was used to generate a standard curve on
mannan-coated wells. Assays were performed in triplicates and
were repeated twice.
2.6. Assay of thrombin-like and factor Xa-like activities
These activities were assayed using previously described
methods [39]. Briefly, 384 well plates were coated with mannan
or IAV as above. After wash, the wells were incubated with various
concentrations of lectins with or without 1% MBL null mouse
serum or 1% MASP-1/3 null mouse serum (MASP source) [40]
diluted in the binding buffer. After wash, wells were incubated
with rhodamine 110-thrombin substrate (R22124, Invitrogen,
Carlsbad, CA) or amino-4-methylcoumarin acetate (AMC)-factor
Xa substrate (222F, American Diagnostica Inc., Stamford, CT) and
read at 500 nm excitation/520 nm emission or 360 nm excitation/
440 nm emission, respectively, using SpectraMax M5. The results
were expressed as arbitrary units (AUs). Assays were performed in
triplicates and were repeated twice.
2.7. Viral neutralizing assay
The assay was performed as previously described [41]. Briefly,
viruses were pre-incubated with lectins and washed and then
incubated with Madin-Darby Canine Kidney (MDCK) cells. Infec-
tion was assayed by FITC-conjugated anti-IAV antibody (Ab)
(Millipore, Billerica, MA). Virus neutralizing activity as %inhibition
W.-C. Chang et al./Biochemical Pharmacology 81 (2011) 388–395
389

Page 3

was calculated by the formula: FFC in test samples/FFC in saline
control ? 100. Data from 5 experiments were combined.
2.8. HA and HA inhibition assay
The assay was previously described [42]. Briefly, virus was
incubated with lectins at various concentrations and assayed for HA
titer using human type O, Rh?RBCs. HA inhibition of IAV by
recombinant lectins at 2mg/ml was measured in round-bottom 96-
well plates (Serocluster U-Vinyl plates; Costar, Cambridge, MA). HA
inhibition was detected as the formation of a RBC pellet. To enable
graphical comparisons of HA inhibition, data were mathematically
converted and expressed as the numberof HAunits inhibited by the
lectins. Four experiments were combined.
2.9. NA inhibition assay
The assay was performed using Amplex Red Neuraminidase
Assay kit (A22178, Molecular Probes, CA), according to the
manufacture’s instructions. IAV (500 U/ml) was mixed with
recombinant lectins at 2 mg/ml and Amplex Red reagents in
40 ml reaction volume. Assay was performed in triplicate. Reaction
was read at 530 nm excitation/590 nm emission using SpectraMax
M5. The results were expressed as %inhibition calculated by the
formula: [(IAV ? (IAV + recombinant lectins)) ? 100]/IAV.
2.10. Viral aggregation assay
The assay was previously described [43]. Briefly, at time 0, viral
suspensions were mixed with 800 ng/ml of lectins in PBS++in a final
volume of 1.0 ml. Under continuous stirring, the light transmission
wasmonitoredatexcitation/emission350 nmduring12 minusingan
SLM/Aminco8000C(SLMInstrument,Urbana,IL)spectrofluorometer.
A decline in light transmission correlates with viral aggregation.
Results are expressed as percent of control light transmission (virus
without lectins). Six experiments were combined.
2.11. Statistical analysis
All data were analyzed by ANOVA or Wilcoxon/Kruskal–Wallis
tests depending on data distribution using JMP software (SAS
institute Inc., Cary, NC). p values less than 0.05 was considered to
be significant.
3. Results
3.1. RCLs bind IAV as efficient as rMBL
We first confirmed RCLs’ binding ability to IAV by comparing to
mannan because all RCLs have MBL-CRD, which preferentially
[(Fig._1)TD$FIG]
recognizes mannan [23]. Therefore, mannan was used as a positive
control throughout the investigation. All RCLs bound to mannan at
comparable efficiency as rMBL in a dose dependent manner
(Fig. 1A). Similarly, all RCLs bound to IAV in a dose dependent
manner although RCL1 and RCL3 demonstrated significantly more
efficient binding at 10 ng/ml compared with RCL2 and rMBL
(Fig.1B). Thus, all RCLs bound to mannan and IAV in a dose
response manner, demonstrating that all RCLs were biologically
functional for the target binding activity.
3.2. RCLs activate the lectin complement pathway more efficiently
than rMBL
On mannan, rMBL showed C4 deposition in a dose dependent
manner only when MASPs (MBL null serum) were supplied as
expected (open circles vs. closed circles in Fig. 2A). In contrast, all
RCLs even without MASPs showed significant C4 deposition
activity. C4 deposition activity by RCLs alone was significantly
(p < 0.0001) higher than that of rMBL even at 10 ng/ml (p < 0.05,
0.005and0.0001forRCL2,RCL1andRCL3,respectively).At100 ng/
ml, C4 deposition activity of RCL2 was highest followed by RCL3
andthenRCL1(p < 0.005forbothRCL4vs.RCL3and RCL3vs.RCL1)
(Fig. 2A). Addition of MASPs significantly (p < 0.0001) increased
C4 deposition activity reaching to plateau at 10 ng/ml. All RCLs
demonstrated comparable activities and were significantly stron-
ger than rMBL at 10 ng/ml (Fig. 2A).
On IAV, C4 deposition activity of RCLs alone was also observed
while it was not detectable by rMBL alone (Fig. 2B). Similar to their
activity on mannan, all RCLs alone activated C4 even at 1 ng/ml
(p < 0.005 and 0.05 for RCL2 and RCL 3, respectively and no
significance for RCL 3) (Fig. 2B). Once again, addition of MASP
significantly (p < 0.0001) enhanced the C4 deposition activity even
at 1 ng/ml on IAV, unlike on mannan (Fig. 2A vs. B) and reached to
plateau at 10 ng/ml. At 1 ng/ml, C4 deposition activity by RCLs with
MASPs was significantly (p < 0.005) stronger than rMBL/MASP
complex (Fig. 2B).
Taken together, these results demonstrated that RCLs were
more efficient in activating the lectin pathway mediated C4
deposition on both mannan and IAV.
3.3. rMBL strongly activates coagulation enzyme-like activities
compared with all RCLs
On mannan, thrombin-like activity was observed in a dose
dependent manner by rMBL when MASP was supplemented as
expected (Fig. 3A). Unlike C4 deposition activity, no thrombin-like
activity was detected without MASP by any recombinant lectin on
both mannan and IAV (Fig.3A and B). Even when MASP was
supplied, only MBL mediated significant thrombin-like activity on
mannan (Fig. 3A). In contrast, all recombinant lectins activated
Fig. 1. Lectin binding assay to mannan (A) or IAV (B). Bound lectins on mannan (control) or influenza A virus (IAV) were detected by monoclonal anti-MBL antibody, which
detected all RCLs. Open circles, rMBL; open triangles, RCL1; open squares, RCL2; open diamonds, RCL3. Data were expressed as mean ? SD, all of which were smaller than sizes
of symbols. *p < 0.05.
W.-C. Chang et al./Biochemical Pharmacology 81 (2011) 388–395
390

Page 4

[(Fig._2)TD$FIG]
Fig. 2.The lectin complementpathway activation activity on mannan (A)orIVA (B). Thelectin pathway activity was assayed asC4 deposition (U/ml)as described in Section2.
Open circles, rMBL; open triangles, RCL1; open squares, RCL2; open diamonds, RCL3. Closed symbols were with MBL null sera to supply MASP. C4 deposition activities (U/ml)
were expressed as mean ? SE, most of which were smaller than sizes of symbols. *p < 0.05; ***p < 0.0001.
[(Fig._3)TD$FIG]
Fig. 3. Thrombin-like activities on mannan (A) or IAV (B) and FXa-like activities on mannan (C) or IAV (D). These activities were assayed using enzyme-specific peptide
substrates, which become fluorescent upon enzymatic digestion. Open circles, rMBL; open triangles, RCL1; open squares, RCL2; open diamonds, RCL3. Closed symbols were
with MBL null sera to supply MASP. Activities (arbitrary units, AUs) were expressed as mean ? SE, most of which were smaller than sizes of symbols. *p < 0.05; ***p < 0.0001.
W.-C. Chang et al./Biochemical Pharmacology 81 (2011) 388–395
391

Page 5

thrombin-like activity on IAV at 1 and 10mg/ml. rMBL-mediated
activity was significantly stronger than that of RCLs (Fig. 3B).
Similartothrombin-likeactivity,whenMASPwassuppliedrMBL
and RCL2 significantly activated FXa-like activity on mannan
compared with RCL1 and RCL3. However, RCL2 showed the activity
only at 10mg/ml while rMBL was significantly active even at 1mg/
ml (Fig. 3C). Once again, all recombinant lectins activated FXa-like
activity on IAV at as low as 0.1mg/ml when MASP was supplied.
However, MBL-mediated activity was significantly stronger than
that of RCLs (Fig. 3D).
For both thrombin-like and FXa-like activities, when MASP-1/3
null mouse serum (MASP-2/sMAP sufficient) was used as MASP
source these activities were abrogated to nearly undetectable
levels as this was expected from our previous study (Fig. 4) [39].
3.4. RCLs demonstrate efficient anti-IAV biologic functions
All RCLs inhibited MDCK cell infection in a dose dependent
manner. The activity was comparable to rMBL (Fig. 5). IC50s (mg/
ml ? SE) for RCL1, RCL2, RCL3, and rMBL were 1.12 ? 0.15, 0.72 ? 0.15,
[(Fig._5)TD$FIG]
1.09 ? 0.14,and0.81 ? 0.16,respectively.Thus,alllectinswereactivein
inhibiting IAV infection to MDCK epithelial cells.
In IAV aggregation activity, RCL2 exceeded all other RCLs and
rMBL (Fig. 6). IAV aggregation by RCL2 was detected as early as
50 s into incubation (Fig. 6). rMBL aggregated IAV better than both
RCL1 and RCL3 after incubation time at 500 and 700 s (p < 0.05 for
both time points, Fig. 6).
RCLs inhibited HA titers as potent as rMBL. Similar to viral
aggregation results, RCL2significantly(p < 0.001) reduced HA titer
at 5 mg/ml compared with other RCLs and rMBL (Fig. 7A). RCL2
abolished HA titer at 10 mg/ml while RCL1, RCL3 and rMBL
required 20 mg/ml to achieve the same effect (Fig. 7A).
As complementing lectins ability to reduce HA titer, HA
inhibition was assayed at 2 mg/ml for all chimeric lectins. The
results showed that all RCLs significantly inhibited HA compared
with rMBL (Fig. 7B), demonstrating that all RCLs were significantly
more efficient in inhibiting IAV-mediated HA.
Another anti-viral activity is inhibition of NA. RCL2 and RCL3
significantlyinhibitedNAcomparedwithRCL1andrMBL,thelatter
twoof whichshowedsimilarNA inhibitoryactivity(Fig. 7C).These
results suggested that RCL2 and RCL3 were significantly effective
in inactivating NA. Taken together, these results further suggest
that RCLs, in particular, RCL2 was effective in preventing IAV
infection into host cells.
4. Discussion
In this study, we have demonstrated that novel RCLs, in
particular,RCL2,isanefficientanti-IAVagent aswesummarizeour
findings in Table 1. RCLs have been generated by replacing various
portions of the collagenous domain of MBL with that of L-FCN, thus
utilizingMBL-CRDfortargetrecognition.MBL-CRDischosenbased
on our understanding that MBL-CRD has broader target recogni-
tion than L-FCN, which preferentially recognizes acetylated
compounds [44]. These results support our recent findings that
these RCLs have a better binding activity against Ebola, Nipah and
Hendra viruses [23]. Taken together, these results demonstrate
that novel RCLs have broad and potent anti-viral activities
compared with its parent rMBL.
Binding ability of all RCLs on mannan (a positive control) was in
a dose dependent manner and was comparable to that of rMBL.
Thus, RCLs maintain MBL-CRD binding activity and introduction of
the collagenous domain of L-FCN does not alter the MBL-CRD
function. Similar to on mannan, RCLs bind IAV in a dose dependent
manner at comparable levels to rMBL, confirming that MBL-CRD is
functional in binding to IAV.
[(Fig._6)TD$FIG]
[(Fig._4)TD$FIG]
Fig. 4. MASP-1/3 dependent thrombin-like (A) and FXa-like (B) activities. Assays
wereperformedasinFig. 3.Lectinswereusedat1 mg/mlandmixedwith1%ofMBL
null or MASP-1/3 null mouse sera. Data were expressed as mean ? SE, mostofwhich
were smaller than sizes of symbols. The data shown was the representative result of
three experiments.
Fig. 5. IAV neutralizing assay. Neutralizing activity was assayed as %inhibition of
viral infection to MDCK cells by comparing lectin-pretreatment to no lectin-
pretreatment. Open circles, rMBL; open triangles, RCL1; open squares, RCL2; open
diamonds, RCL3. Results were expressed as mean ? SE of %inhibition. *p < 0.05.
Fig.6.Viralaggregationassaybylectins.Viralaggregationbylectinswas assayedby
the reduction of light transmittance and expressed as % of control (saline) light
transmittance.Datawere expressed
***p < 0.0001.
asmean ? SE.*p < 0.05;**p < 0.005;
W.-C. Chang et al./Biochemical Pharmacology 81 (2011) 388–395
392

Page 6

MBL in a complex with MASPs initiates the lectin complement
pathway, one of the key functions of MBL. By replacing various
portions of collagen domain with that of L-ficolin, all RCLs without
MASP activate the lectin complement pathway, which is also
observed by rMBL but requires 10 mg/ml, 1000-fold more protein
(unpublished observation). Importantly, RCL-mediated lectin
complement pathway activation activity is augmented by MASP
supplementation, suggesting that the lectin complement pathway
activationactivitywould beefficiently augmentedinvivo.We have
previously shown that the lectin complement pathway activation
activity correlates with host protection from bacterial infection,
including S. aureus infection in vivo [36,45]. Taken together, these
observations suggest that RCLs are efficient activators of the lectin
complement pathway and could be administered with smaller
dose than rMBL. Further investigation is required to determine in
vivo efficiency.
AlthoughSP-AandSP-Dbelongtothecollectinfamilyandshare
structural and functional similarity with MBL and FCNs these
surfactantproteins do not activate the lectincomplementpathway
[46]. SP-A rather inhibits complement activation [47]. The
chimeric protein of SP-D collagenous domain and MBL-CRD has
increased anti-viral characteristics compared with its parents [34],
however complement activation activity has not been examined.
Interestingly, when Wallis and his colleagues introduced MASP
binding sequence of MBL/FCN into SP-A the chimeric SP-A became
constitutively active in the lectin complement pathway. This is not
the case for MBL/FCN chimera because their activities are dose
dependent and correlate with ligand binding. Nevertheless, superb
lectin complement pathway activities of RCLs may be explained by
the idea that altering MASP-binding region might have allowed to
increase MASP-2 binding to RCLs and also MASP-2 activity itself
because the lectin complement pathway is predominantly
mediated by MASP-2 [24].
Conversely, compared with rMBL, all RCLs now have reduced
thrombin-like and FXa-like coagulation enzyme activities, which
are mediated by MASP-1/3. MASP-1 has been linked to coagulation
enzyme-like activities as we also have demonstrated in this study,
confirming previous findings of our own and others [28,39]. One
can speculate that while all RCLs efficiently bind to MASP-2 their
MASP-1 binding is diminished either by competition or by
structural conformational change. Another possibility is due to
triggering of different amplification loops. It has been proposed
that the alternative pathway amplifies the classical pathway and
the lectin pathway [48–50]. There could be other such amplifica-
tion loop as it has become clear that interaction of complement
pathway and coagulation pathway is more complex. Further
detailed investigations are required to dissect out such pathways
and cascades.
The reduced coagulation enzyme-like activities of RCLs are
advantageousoverrMBLfor thefollowingreasons:Bloodfluidityis
maintained on a fine balance of clotting (coagulation/thrombosis)
and bleeding, thus, tip over to one side would trigger coagulation
disorders, including disseminated intravascular coagulation [51].
So far, a phase 1 clinical study of rMBL did not result in adverse
effect in healthy volunteers [52]. Taken together, our results
suggest that RCLs have low coagulation enzyme-like activities,
such as thrombin-like and FXa-like activities. This should be
advantageous in clinical use, as it would have less risk to trigger
coagulation disorder and related side effects.
MBL has been known to neutralize virus as this protein was
initially discovered as a serum factor, the b-inhibitor of IAV in
1940s and re-identified to be MBL in 1990s [19]. Viral
neutralization is assayed by infectivity of lectin-treated virus to
MDCK cells, thus the activity indicates integrity of virus [41]. All
RCLs neutralize IAV at similar level to rMBL, suggesting that all
RCLs are at least functionally as efficient as rMBL in vitro. Further
investigations would require assessing these activities in more
physiological way, in which primary cells would be used in the
presence of serum as well as in vivo infection studies.
Further characterization has revealed that these RCLs, in
particular, RCL2 surpasses rMBL for important aspects of anti-
Table 1
Summary of RCL activities.
Activities Recombinant lectins
RCL1 RCL2RCL3 rMBL
Binding
LCP activation activity
Thrombin-like activity
FXa-like activity
Viral aggregation
Reduction of HA titer
HA inhibition
NA inhibition
Viral infectivity (in vitro)
+
++
+
+
+
+
++
+
+
+
+++
+
+
+++
++
++
+++
+
+
++
+
+
+
+
++
++
+
+
+
+++
+++
+
+
+
+
+
Note: LCP, lectin complement pathway; FXa, factor X-activated; HA, hemaggluti-
nation;NA,neuraminidase.Activities werescored asfollows: +,positive;++, strong;
+++, very strong.
[(Fig._7)TD$FIG]
Fig. 7. HA titers and HA inhibition (HAI) and NA inhibition (NAI) by lectins. (A) HA
titers were inhibited by lectins. Closed circles, rMBL; closed triangles, RCL1, closed
squares, RCL2; closed diamonds, RCL3. The results were expressed as mean ? SE.(B)
HAI assay complemented lectin’s effects on HA titers in (A). Bars indicated mean ? SE.
*p < 0.05 against rMBL. (C) NAI assay examined lectin’s effect on viral NA activity. Bars
indicated mean ? SE. **p < 0.005; *p < 0.05. Both statistics were against RCL1 and
rMBL.
W.-C. Chang et al./Biochemical Pharmacology 81 (2011) 388–395
393

Page 7

IAV functions. For example, aggregation of virus is an efficientanti-
viral mechanism as aggregated virus loses infectivity [20]. In
regard to this activity, RCL2 surpassed all other lectins, including
other two RCLs. RCL2’s superb anti-viral activity is also evident in
reduction of HA titer. RCL2’s activity is further confirmed by HA
inhibition, which is an important anti-viral activity as it suggests
that RCL2 would inhibit viral attachment to the host cells [20],
thereby preventing infection. Similarly, RCL2 inhibits NA, suggest-
ing that RCL inactivate NA that digests sialic acid, which is
expressed on intact host cells, and promotes infection [22]. The
findingconfirmsandextendsthepreviousreportthatMBLnotonly
binds to NA but also inactivates its activity [22]. The RCL2’s
superior activities may be partly attributed to significantly higher
binding affinity to mannan and GlcNAc, another favored MBL-
ligand, compared with other lectins [23].
In conclusion, our results demonstrate that RCLs, in particu-
lar, RCL2 are more potent anti-IAV reagents than MBL as
summarized in Table 1. We also speculate that these RCLs would
recognize infected host cells that may be expressing HA and NA.
Our studies also provide new insights into understanding how
collagenous domains would contribute to biologic functions,
such as complement activation and coagulation, which are
mediated by MASPs. Further in vivo investigation is encouraged
to examine their efficiency as a new innate immune therapeutics
against IAV infections.
Conflict of interest
All authors have no financial conflict.
Acknowledgements
We would like to thank Enzon pharmaceuticals for providing
rMBL. We also thank Dr. Gregory Stahl, Center for Experimental
Therapeutics and Reperfusion Injury, Department of Anesthesiol-
ogy, Perioperative and Pain Medicine, Brigham and Women’s
Hospital, Harvard Medical School, Boston, MA 02115, for providing
mouse anti-human MBL monoclonal antibody (2A9). This work
was supported by NIH grants UO1 AI074503-01 and R21
AI077081-01A1 (KT) and UO1 AI070330-01 (EVS).
References
[1] Nayak DP, Hui EK, Barman S. Assembly and budding of influenza virus. Virus
Res 2004;106:147–65.
[2] Lynch 3rd JP, Walsh EE. Influenza: evolving strategies in treatment and
prevention. Semin Respir Crit Care Med 2007;28:144–58.
[3] Munoz FM. Influenza virus infection in infancy and early childhood. Paediatr
Respir Rev 2003;4:99–104.
[4] Bouree P. Immunity and immunization in elderly. Pathol Biol (Paris)
2003;51:581–5.
[5] Saito R, Sato I, Suzuki Y, Baranovich T, Matsuda R, Ishitani N, et al. Reduced
effectiveness of oseltamivir in children infected with oseltamivir-resistant
influenza A (H1N1) viruses with His275Tyr mutation. Pediatr Infect Dis J 2010.
[6] HoffmannJA,KafatosFC,JanewayCA,EzekowitzRA.Phylogeneticperspectives
in innate immunity. Science 1999;284:1313–8.
[7] Uemura K, Saka M, Nakagawa T, Kawasaki N, Thiel S, Jensenius JC, et al. L-MBP
is expressed in epithelial cells of mouse small intestine. J Immunol
2002;169:6945–50.
[8] Hansen S, Thiel S, Willis A, Holmskov U, Jensenius JC. Purification and charac-
terization of two mannan-binding lectins from mouse serum. J Immunol
2000;164:2610–8.
[9] Oka S, Ikeda K, Kawasaki T, Yamashina I. Isolation and characterization of two
distinct mannan-binding proteins from rat serum. Arch Biochem Biophys
1988;260:257–66.
[10] Weis WI, Drickamer K. Trimeric structure of a C-type mannose-binding
protein. Structure 1994;2:1227–40.
[11] Holmskov U, Malhotra R, Sim RB, Jensenius JC. Collectins: collagenous C-type
lectinsof the innate immunedefense system.Immunol Today 1994;15:67–74.
[12] Crouch E, Hartshorn K, Ofek I. Collectins and pulmonary innate immunity.
Immunol Rev 2000;173:52–65.
[13] Pastva AM, Wright JR, Williams KL. Immunomodulatory roles of surfactant
proteins A and D: implications in lung disease. Proc Am Thorac Soc
2007;4:252–7.
[14] LeVine AM, Whitsett JA, Hartshorn KL, Crouch EC, Korfhagen TR. Surfactant
protein D enhances clearance of influenza A virus from the lung in vivo. J
Immunol 2001;167:5868–73.
[15] Crouch E, Hartshorn K, Horlacher T, McDonald B, Smith K, Cafarella T, et al.
Recognition of mannosylated ligands and influenza A virus by human surfac-
tant protein D: contributions of an extended site and residue 343. Biochemis-
try 2009;48:3335–45.
[16] Hartshorn KL, Webby R, White MR, Tecle T, Pan C, Boucher S, et al. Role of viral
hemagglutinin glycosylation in anti-influenza activities of recombinant sur-
factant protein D. Respir Res 2008;9:65.
[17] Tecle T, White MR, Sorensen G, Gantz D, Kacak N, Holmskov U, et al. Critical
role for cross-linking of trimeric lectin domains of surfactant protein D in
antiviral activity against influenza A virus. Biochem J 2008;412:323–9.
[18] Li G, Siddiqui J, Hendry M, Akiyama J, Edmondson J, Brown C, et al. Surfactant
protein-A – deficient mice display an exaggerated early inflammatory re-
sponse to a beta-resistant strain of influenza A virus. Am J Respir Cell Mol Biol
2002;26:277–82.
[19] Anders EM,Hartley CA,JacksonDC. Bovine andmouse serumbeta inhibitors of
influenza A viruses are mannose-binding lectins. Proc Natl Acad Sci U S A
1990;87:4485–9.
[20] Hartshorn KL, Sastry K, Brown D, White MR, Okarma TB, Lee YM, et al.
Conglutinin acts as an opsonin for influenza A viruses. J Immunol
1993;151:6265–73.
[21] ReadingPC,MoreyLS,CrouchEC,AndersEM.Collectin-mediatedantiviralhost
defense of the lung: evidence from influenza virus infection of mice. J Virol
1997;71:8204–12.
[22] Kase T, Suzuki Y, Kawai T, Sakamoto T, Ohtani K, Eda S, et al. Human mannan-
binding lectin inhibits the infection of influenza A virus without complement.
Immunology 1999;97:385–92.
[23] Michelow IC, Dong M, Mungall BA, Yantosca LM, Lear C, Ji X, et al. A novel L-
ficolin/mannose-binding lectin chimeric molecule with enhanced activity
against Ebola virus. J Biol Chem 2010;285:24729–3.
[24] Takahashi M,MoriS,Shigeta S,FujitaT.Roleof MBL-associatedserineprotease
(MASP) on activation of the lectin complement pathway. Adv Exp Med Biol
2007;598:93–104.
[25] Thiel S, Vorup-Jensen T, Stover CM, Schwaeble W, Laursen SB, Poulsen K, et al.
Asecondserineproteaseassociatedwithmannan-bindinglectinthatactivates
complement. Nature 1997;386:506–10.
[26] Matsushita M, Fujita T. Activation of the classical complement pathway by
mannose-binding protein in association with a novel C1s-like serine protease.
J Exp Med 1992;176:1497–502.
[27] Krarup A, Wallis R, Presanis JS, Gal P, Sim RB. Simultaneous activation of
complement and coagulation by MBL-associated serine protease 2. PLoS ONE
2007;2:e623.
[28] Presanis JS, Hajela K, Ambrus G, Gal P, Sim RB. Differential substrate and
inhibitorprofilesfor human MASP-1
2004;40:921–9.
[29] Muta T, Iwanaga S. Clotting and immune defense in Limulidae. Prog Mol
Subcell Biol 1996;15:154–89.
[30] Endo Y, Takahashi M, Fujita T. Lectin complement system and pattern recog-
nition. Immunobiology 2006;211:283–93.
[31] Honore C, Rorvig S, Hummelshoj T, Skjoedt MO, Borregaard N, Garred P.
Tethering of Ficolin-1 to cell surfaces through recognition of sialic acid by
the fibrinogen-like domain. J Leukocyte Biol 2010.
[32] Endo Y, Nakazawa N, Liu Y, Iwaki D, Takahashi M, Fujita T, et al. Carbohydrate-
bindingspecificitiesofmouseficolinA,asplicingvariantofficolinAandficolin
B and their complex formation with MASP-2 and sMAP. Immunogenetics
2005;57:837–44.
[33] Takahashi K, Ezekowitz RA. The role of the mannose-binding lectin in innate
immunity. Clin Infect Dis 2005;41(Suppl. 7):S440–4.
[34] White MR, Crouch E, Chang D, Sastry K, Guo N, Engelich G, et al. Enhanced
antiviral and opsonic activity of a human mannose-binding lectin and surfac-
tant protein D chimera. J Immunol 2000;165:2108–15.
[35] Hartshorn KL, Collamer M, Auerbach M, Myers JB, Pavlotsky N, Tauber AI.
Effects of influenza A virus on human neutrophil calcium metabolism. J
Immunol 1988;141:1295–301.
[36] Shi L, Takahashi K, Dundee J, Shahroor-Karni S, Thiel S, Jensenius JC, et al.
Mannose-binding lectin-deficient mice are susceptible to infection with
Staphylococcus aureus. J Exp Med 2004;199:1379–90.
[37] Collard CD, Montalto MC, Reenstra WR, Buras JA, Stahl GL. Endothelial oxida-
tive stress activates the lectin complement pathway: role of cytokeratin 1. Am
J Pathol 2001;159:1045–54.
[38] Moller-Kristensen M, Hamblin MR, Thiel S, Jensenius JC, Takahashi K. Burn
injury reveals altered phenotype in mannan-binding lectin-deficient mice. J
Invest Dermatol 2007;127:1524–31.
[39] Takahashi K, Chang WC, Takahashi M, Pavlov V, Ishida Y, La Bonte L, et al.
Mannose-binding lectin and its associated proteases (MASPs) mediate coagula-
tionanditsdeficiencyisariskfactorindevelopingcomplicationsfrominfection,
including disseminated intravascular coagulation. Immunobiology 2010.
[40] Takahashi M, Iwaki D, Kanno K, Ishida Y, Xiong J, Matsushita M, et al.
Mannose-binding lectin (MBL)-associated serine protease (MASP)-1 contri-
butes to activation of the lectin complement pathway. J Immunol
2008;180:6132–8.
[41] Hartshorn KL, Sastry KN, Chang D, White MR, Crouch EC. Enhanced anti-
influenza activity of a surfactant protein D and serum conglutinin fusion
protein. Am J Physiol Lung Cell Mol Physiol 2000;278:L90–98.
and MASP-2.Mol Immunol
W.-C. Chang et al./Biochemical Pharmacology 81 (2011) 388–395
394

Page 8

[42] van Eijk M, White MR, Crouch EC, Batenburg JJ, Vaandrager AB, Van Golde LM,
et al. Porcine pulmonary collectins show distinct interactions with influenza A
viruses: role of the N-linked oligosaccharides in the carbohydrate recognition
domain. J Immunol 2003;171:1431–40.
[43] Hartshorn KL, White MR, Shepherd V, Reid K, Jensenius JC, Crouch EC.
Mechanisms of anti influenza activity of surfactant proteins A and D: com-
parison with serum collectins. Am J Physiol 1997;273:L1156–6.
[44] Takahashi K, Ficolins. Encyclopedia of life sciences. Chichester: John Wiley &
Sons Ltd.; 2008.
[45] Moller-Kristensen M, Ip WK, Shi L, Gowda LD, Hamblin MR, Thiel S, et al.
Deficiency of mannose-binding lectin greatly increases susceptibility to post-
burn infection with Pseudomonas aeruginosa. J Immunol 2006;176:1769–75.
[46] Holmskov U, Thiel S, Jensenius JC. Collections and ficolins: humoral lectins of
the innate immune defense. Annu Rev Immunol 2003;21:547–78.
[47] Watford WT, Wright JR, Hester CG, Jiang H, Frank MM. Surfactant protein A
regulates complement activation. J Immunol 2001;167:6593–600.
[48] Banda NK, Takahashi K, Wood AK, Holers VM, Arend WP. Pathogenic comple-
ment activation in collagen antibody-induced arthritis in mice requires am-
plification by the alternative pathway. J Immunol 2007;179:4101–9.
[49] Ganter MT, Brohi K, Cohen MJ, Shaffer LA, Walsh MC, Stahl GL, et al. Role of the
alternative pathway in the early complement activation following major
trauma. Shock 2007;28:29–34.
[50] Brouwer N, Dolman KM, van Zwieten R, Nieuwenhuys E, Hart M, Aarden LA,
et al. Mannan-binding lectin (MBL)-mediated opsonization is enhanced by
the alternative pathway amplification loop. Mol Immunol 2006;43:2051–
60.
[51] Hess JR, Brohi K, Dutton RP, Hauser CJ, Holcomb JB, Kluger Y, et al. The
coagulopathy of trauma: a review of mechanisms. J Trauma-Injury Infect Crit
Care 2008;65:748–54.
[52] Petersen KA, Matthiesen F, Agger T, Kongerslev L, Thiel S, Cornelissen K, et al.
Phase I safety, tolerability, and pharmacokinetic study of recombinant human
mannan-binding lectin. J Clin Immunol 2006;26:465–75.
W.-C. Chang et al./Biochemical Pharmacology 81 (2011) 388–395
395