To the editor:

The use of anti-D immunoglobulin to prevent hemolytic disease of the fetus and newborn (HDFN) is a success of antibody-mediated immune suppression (AMIS) in the clinic. A monoclonal antibody (mAb) to replace donor-derived anti-D would be beneficial to eliminate current issues with anti-D.1,2  There have been several attempts to create replacement monoclonal anti-D therapies, however, no products have been as effective as polyclonal anti-D and some have led to enhanced alloimmunization.3  This highlights the necessity of increased knowledge of anti-D mechanisms.

It has been suggested that immunoglobulin G (IgG) glycosylation on the Fc domain of AMIS antibodies may influence their ability to induce AMIS.3  Removal of the IgG Fc glycans has been associated with impaired Fc receptor interactions, complement activation, and antibody-dependent cellular cytoxicity.4,5  In particular, variations in anti-D IgG Fc fucosylation have been shown to impact HDFN severity in patients with lower levels of fucosylation correlated with enhanced HDFN severity.6  Glycosylation of several alloantibodies in pregnant women has been shown to have antigen-specific differences in glycosylation patterns compared with total IgG.7  Differences in IgG glycosylation of monoclonal anti-D produced in vitro compared with polyclonal anti-D could potentially contribute to the decreased efficacy of monoclonal anti-D for the prevention of red blood cell (RBC) alloimmunization.3  In this study, we completely removed the Fc glycan for 4 anti-RBC antibodies and show that AMIS can occur independently of IgG Fc glycan–dependent RBC clearance in a mouse model of RBC alloimmunization.

This RBC alloimmunization model consists of immunizing C57BL/6 mice (6-8 weeks old, Charles River Laboratories, Kingston, NY) with hen egg lysozyme (HEL), ovalbumin and human Duffy protein (HOD) transgenic RBCs.8  All animal studies were approved by the St. Michael’s Hospital Animal Care Committee. We tested AMIS induction with a panel of 4 anti-RBC mAbs: (1) MIMA 29 (mouse anti-Duffy [Fy3 ]; IgG2a),9  (2) CBC-512 (mouse anti-Duffy [Fy3 ]; IgG1), (3) 4B7 (mouse anti-HEL; IgG1), and (4) 6D7 (mouse anti-HEL; IgG1). To evaluate the role of Fc glycans for IgG AMIS induction, Fc region deglycosylated antibodies were generated by treatment with the enzyme peptide-N-glycosidase F (PNGase F, catalog # P0704S, New England Biolabs, Whitby, ON, Canada), as previously performed in Yu et al.10 

Successful Fc glycan removal for each antibody was confirmed by ultra-performance liquid chromatography (UPLC) (Figure 1A) as well as a band-shift assay and additional UPLC data (supplemental Figure 1A, available on the Blood Web site). The deglycosylated mAbs retained the ability to bind to HOD-RBCs similarly to the wild-type version (supplemental Figure 1B).

Figure 1.

Analysis of wild-type and deglycosylated anti-RBC antibodies and their ability to induce RBC clearance and AMIS. Wild-type and deglycosylated antibodies were first evaluated for the efficiency of glycan removal. Glycans were released from IgG heavy chains by in-gel PNGase F digestion and then labeled with 2-aminoanthranilic acid and analyzed by UPLC. The symbols used for different glycan structures is based on Harvey et al.22  (A) Heavy chains of all samples had detectable glycan structures in the wild-type form of the heavy chains (blue lines), but no detectable structures after deglycosylation (red lines). The loss of the heavy chain signal after the deglycosylation reaction with PNGase F indicates that the Fc glycan removal was successful. (B) The ability of anti-RBC antibodies (MIMA 29 and CBC-512) to induce RBC clearance was analyzed by determining the percentage of surviving PKH26+ HOD-RBCs in the circulation of mice. All mice except the naive treatment group received 108 PKH26-labeled HOD-RBCs IV by tail vein injection. After 24 hours, mice were injected with no antibody (HOD) or 5 μg of each antibody assessed: wild-type MIMA 29 (MIMA 29), deglycosylated MIMA 29 (deMIMA 29), wild-type CBC-512 (CBC-512), or deglycosylated CBC-512 (deCBC-512). The percentage of remaining PKH26+ HOD-RBCs in the circulation was evaluated before (0 hour) and 2, 24, 48, and 72 hours after antibody injection. Mice were bled for serum 7 days after PKH26+ HOD-RBC transfusion and HEL-specific IgM (C) and IgG (D) antibody levels were evaluated by enzyme-linked immunosorbent assay. Data represent individual values from mice from ≥3 separate experiments. Data were expressed as mean ± standard error of the mean and analyzed by 1-way analysis of variance with Tukey’s multiple comparison test. ns, not significant. **P < .01; ***P < .001; ****P < .0001.

Figure 1.

Analysis of wild-type and deglycosylated anti-RBC antibodies and their ability to induce RBC clearance and AMIS. Wild-type and deglycosylated antibodies were first evaluated for the efficiency of glycan removal. Glycans were released from IgG heavy chains by in-gel PNGase F digestion and then labeled with 2-aminoanthranilic acid and analyzed by UPLC. The symbols used for different glycan structures is based on Harvey et al.22  (A) Heavy chains of all samples had detectable glycan structures in the wild-type form of the heavy chains (blue lines), but no detectable structures after deglycosylation (red lines). The loss of the heavy chain signal after the deglycosylation reaction with PNGase F indicates that the Fc glycan removal was successful. (B) The ability of anti-RBC antibodies (MIMA 29 and CBC-512) to induce RBC clearance was analyzed by determining the percentage of surviving PKH26+ HOD-RBCs in the circulation of mice. All mice except the naive treatment group received 108 PKH26-labeled HOD-RBCs IV by tail vein injection. After 24 hours, mice were injected with no antibody (HOD) or 5 μg of each antibody assessed: wild-type MIMA 29 (MIMA 29), deglycosylated MIMA 29 (deMIMA 29), wild-type CBC-512 (CBC-512), or deglycosylated CBC-512 (deCBC-512). The percentage of remaining PKH26+ HOD-RBCs in the circulation was evaluated before (0 hour) and 2, 24, 48, and 72 hours after antibody injection. Mice were bled for serum 7 days after PKH26+ HOD-RBC transfusion and HEL-specific IgM (C) and IgG (D) antibody levels were evaluated by enzyme-linked immunosorbent assay. Data represent individual values from mice from ≥3 separate experiments. Data were expressed as mean ± standard error of the mean and analyzed by 1-way analysis of variance with Tukey’s multiple comparison test. ns, not significant. **P < .01; ***P < .001; ****P < .0001.

Close modal

The ability of anti-D to prevent HDFN has long been thought to be due to IgG interactions with Fc receptors on phagocytic cells, leading to rapid clearance of the erythrocytes.11  IgG interactions with Fc receptors after removal of the conserved Fc N-linked glycan are impaired.4  The 2 Duffy-specific mAbs (MIMA 29; IgG2a and CBC-512; IgG1) used in this study induce clearance of HOD-RBCs in vivo.12,13  To study RBC clearance, HOD-RBCs were labeled with the fluorescent dye PKH26 to track the transfused cells ex vivo.14-16  C57BL/6 mice were transfused with 108 PKH26-labeled HOD-RBCs (PKH26+ HOD-RBCs), and 24 hours later, mice were injected IV with 5 μg of wild-type or deglycosylated variants of MIMA 29 or CBC-512, or phosphate-buffered saline. Peripheral blood was collected before (0 hours) and 2 hours, 24 hours, 48 hours, and 72 hours after antibody injection to assess the ex vivo clearance kinetics of PKH26+ HOD-RBCs by flow cytometry (Figure 1B). Complete deglycosylation of the MIMA 29 antibody partially impacted its ability to clear HOD-RBCs (Figure 1B), whereas deglycosylation of the CBC-512 antibody completely impaired its ability to clear the transfused HOD-RBCs (Figure 1B). This data suggests that CBC-512 clearance is fully dependent on Fc region glycans, whereas MIMA 29 clearance is only partially dependent on Fc glycans.

To analyze the impact of deglycosylation on AMIS induction, sera from recipient mice were collected 7 days after injection of PKH26+ HOD-RBCs and the HEL-specific IgM and IgG response analyzed by enzyme-linked immunosorbent assay.12,17  Mice that were transfused with HOD-RBCs and injected with wild-type MIMA 29 (AMIS conditions) had minimal IgM and IgG anti-HEL responses (Figure 1C-D), whereas mice that received deglycosylated MIMA 29 displayed less AMIS activity (Figure 1C-D). Therefore, the complete removal of the Fc glycan from MIMA 29 partially reduced AMIS activity and partially impacted PKH26+ HOD-RBCs clearance in vivo. In contrast, the deglycosylated or wild-type version of CBC-512 both completely suppressed the IgM and IgG anti-HEL response to the transfused HOD-RBCs (Figure 1C-D). Thus, complete removal of the Fc glycan from CBC-512 did not prevent the complete AMIS effect mediated by this antibody. The fact that deglycosylated CBC-512 was able to induce complete AMIS despite having no ability to clear the HOD-RBCs provides direct evidence that AMIS can occur independently of IgG Fc glycan–dependent clearance of RBCs in the HOD model of RBC alloimmunization. This finding challenges the necessity for clearance of allogeneic RBCs for successful AMIS induction, which is considered important with anti-D.18,19 

In addition, 2 monoclonal anti-HEL–specific antibodies (6D7; IgG1 and 4B7; IgG1) that do not cause erythrocyte clearance, but are known to mediate AMIS activity12,17  were fully deglycosylated, and despite complete Fc deglycosylation (Figure 1A), retained normal AMIS activity (supplemental Figure 1E-F). These data together show that removal of the Fc glycan did not critically impact AMIS induction for three-fourths of tested anti-RBC antibodies. The one antibody that was partially impacted by Fc glycan removal (MIMA 29) was still able to mediate a partial AMIS effect despite the complete absence of any detectable glycan. It is possible that this deglycosylated IgG2a antibody may retain some residual Fc receptor binding activity, as studies using endoglycosidase S (which removes the majority of the IgG Fc glycan) results in an IgG2a with some Fc receptor I and IV binding activity.20,21  Whether the complete removal of the glycan (as performed in this study) better abrogates Fc receptor binding is uncertain. Interestingly, the removal of the glycan from the MIMA 29 antibody also partially impacted its ability to clear RBCs, and it remains possible that with this antibody, RBC clearance may play a partial role in AMIS activity. Table 1 summarizes the characteristics of the panel of anti-RBC antibodies evaluated in this study and the impact of Fc glycan removal on AMIS induction.

Table 1.

Anti-RBC antibody characteristics and AMIS induction

AntibodyIsotypeReactivityRBC clearance*Antigen modulationAMIS
4B7 IgG1 HEL — No Yes 
de4B7 IgG1 (deglycosylated) HEL  No Yes 
6D7 IgG1 HEL Unknown§ Unknown Yes 
de6D7 IgG1 (deglycosylated) HEL Unknown§ Unknown Yes 
MIMA 29 IgG2a Duffy +++ Yes Yes 
deMIMA 29 IgG2a (deglycosylated) Duffy ++ Unknown Yes (partial) 
CBC-512 IgG1 Duffy +++ Unknown Yes 
deCBC-512 IgG1 (deglycosylated) Duffy — Unknown Yes 
AntibodyIsotypeReactivityRBC clearance*Antigen modulationAMIS
4B7 IgG1 HEL — No Yes 
de4B7 IgG1 (deglycosylated) HEL  No Yes 
6D7 IgG1 HEL Unknown§ Unknown Yes 
de6D7 IgG1 (deglycosylated) HEL Unknown§ Unknown Yes 
MIMA 29 IgG2a Duffy +++ Yes Yes 
deMIMA 29 IgG2a (deglycosylated) Duffy ++ Unknown Yes (partial) 
CBC-512 IgG1 Duffy +++ Unknown Yes 
deCBC-512 IgG1 (deglycosylated) Duffy — Unknown Yes 
*

RBC clearance determined by the percentage of cells remaining 2 hours after injection. +++, <50%; ++, <75%; –, >90%.

Antigen modulation (also known as antigen loss) is a process by which an erythrocyte antigen becomes weakened or undetectable subsequent to antibody binding. The data for antigen modulation are based on previously published data by others.8,13 

We presume that de4B7 does not induce RBC clearance or antigen modulation because the wild-type 4B7 does not.

§

Although we did not test RBC clearance by 6D7, all monoclonal and polyclonal anti-HEL antibodies tested thus far have not induced RBC clearance.12 

This work demonstrates that the ability of RBC-specific mAbs to prevent RBC alloimmunization is not critically dependent on the presence of IgG Fc glycans. However, the differential AMIS effect seen for one mAb (MIMA 29) highlights the complexity of AMIS and provides support for the notion that several mechanisms may be involved in the ability of antibodies to mediate AMIS. This work is important for informing and guiding future developments in monoclonal efforts to replace polyclonal anti-D. This information also suggests that a shift in thinking for the criteria used to develop monoclonal anti-D therapies may be necessary, because past attempts have always emphasized the RBC clearance abilities of the mAbs.3 

In summary, our data demonstrate that AMIS can be successfully achieved in the absence of IgG Fc glycan–dependent clearance of transfused allogeneic RBCs.

The online version of this article contains a data supplement.

Acknowledgments: The authors thank James Zimring for providing the transgenic HOD mouse model and HEL-specific mAbs (4B7, IgG1, 6D7, and IgG1). The authors also thank Marion Reid and Gregory Halverson for their donation of MIMA 29. The authors also thank our colleagues Andrew Crow, Joan Legarda, Alaa Amash, Danila Leontyev, Peter Norris, Ramsha Khan, Yawen Wang and the St. Michael’s Hospital Research Vivarium staff.

This work was supported by grant CBS221511 from Health Canada as part of the Canadian Blood Services/Canadian Institutes of Health Research (CBS/CIHR) partnership fund as well as an Intramural grant from the Canadian Blood Services (A.H.L.).

The views expressed herein do not necessarily represent the view of the federal government of Canada.

Contribution: D.M. designed and performed the experiments, analyzed the data, and wrote the manuscript; Y.C.-L. performed the experiments and analyzed the data; L.B. analyzed data; N.P.L.L. and M.C. performed glycan analysis; X.Y. analyzed the data; M.U. contributed antibodies; and A.H.L. designed the research, analyzed the data, provided grant funding, and wrote the manuscript; and all authors commented on and approved the manuscript.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Alan H. Lazarus, Keenan Research Centre, St. Michael's Hospital, 30 Bond St, Toronto, ON M5B 1W8, Canada; e-mail: lazarusa@smh.ca.

1.
Urbaniak
SJ
,
Greiss
MA
.
RhD haemolytic disease of the fetus and the newborn
.
Blood Rev
.
2000
;
14
(
1
):
44
-
61
.
2.
Qureshi
H
,
Massey
E
,
Kirwan
D
, et al
British Society for Haematology
.
BCSH guideline for the use of anti-D immunoglobulin for the prevention of haemolytic disease of the fetus and newborn
.
Transfus Med
.
2014
;
24
(
1
):
8
-
20
.
3.
Kumpel
BM
.
Lessons learnt from many years of experience using anti-D in humans for prevention of RhD immunization and haemolytic disease of the fetus and newborn
.
Clin Exp Immunol
.
2008
;
154
(
1
):
1
-
5
.
4.
Nose
M
,
Wigzell
H
.
Biological significance of carbohydrate chains on monoclonal antibodies
.
Proc Natl Acad Sci USA
.
1983
;
80
(
21
):
6632
-
6636
.
5.
Duncan
AR
,
Winter
G
.
The binding site for C1q on IgG
.
Nature
.
1988
;
332
(
6166
):
738
-
740
.
6.
Kapur
R
,
Della Valle
L
,
Sonneveld
M
, et al
.
Low anti-RhD IgG-Fc-fucosylation in pregnancy: a new variable predicting severity in haemolytic disease of the fetus and newborn
.
Br J Haematol
.
2014
;
166
(
6
):
936
-
945
.
7.
Sonneveld
ME
,
Koelewijn
J
,
de Haas
M
, et al
.
Antigen specificity determines anti-red blood cell IgG-Fc alloantibody glycosylation and thereby severity of haemolytic disease of the fetus and newborn
.
Br J Haematol
.
2017
;
176
(
4
):
651
-
660
.
8.
Zimring
JC
,
Cadwell
CM
,
Chadwick
TE
, et al
.
Nonhemolytic antigen loss from red blood cells requires cooperative binding of multiple antibodies recognizing different epitopes
.
Blood
.
2007
;
110
(
6
):
2201
-
2208
.
9.
Wasniowska
K
,
Lisowska
E
,
Halverson
GR
,
Chaudhuri
A
,
Reid
ME
.
The Fya, Fy6 and Fy3 epitopes of the Duffy blood group system recognized by new monoclonal antibodies: identification of a linear Fy3 epitope
.
Br J Haematol
.
2004
;
124
(
1
):
118
-
122
.
10.
Yu
X
,
Menard
M
,
Seabright
G
,
Crispin
M
,
Lazarus
AH
.
A monoclonal antibody with anti-D-like activity in murine immune thrombocytopenia requires Fc domain function for immune thrombocytopenia ameliorative effects
.
Transfusion
.
2015
;
55
(
6 Pt 2
):
1501
-
1511
.
11.
Marjoram
D
,
Cruz-Leal
Y
,
Bernardo
L
,
Lazarus
AH
.
A role for red cell clearance in antibody-mediated inhibition of erythrocyte alloimmunization?
ISBT Sci Ser
.
2017
;
12
(
1
):
196
-
201
.
12.
Yu
H
,
Stowell
SR
,
Bernardo
L
, et al
.
Antibody-mediated immune suppression of erythrocyte alloimmunization can occur independently from red cell clearance or epitope masking in a murine model
.
J Immunol
.
2014
;
193
(
6
):
2902
-
2910
.
13.
Stowell
SR
,
Liepkalns
JS
,
Hendrickson
JE
, et al
.
Antigen modulation confers protection to red blood cells from antibody through Fcγ receptor ligation
.
J Immunol
.
2013
;
191
(
10
):
5013
-
5025
.
14.
Chapanian
R
,
Constantinescu
I
,
Brooks
DE
,
Scott
MD
,
Kizhakkedathu
JN
.
In vivo circulation, clearance, and biodistribution of polyglycerol grafted functional red blood cells
.
Biomaterials
.
2012
;
33
(
10
):
3047
-
3057
.
15.
Xue
G
,
Li
J
,
Liu
J
,
Wu
H
,
Hou
Y
.
[In vivo tracking of PKH26-labeled human umbilical cord mesenchymal stem cells after transplantation into rats with liver cirrhosis]
.
Zhonghua Gan Zang Bing Za Zhi
.
2014
;
22
(
12
):
910
-
914
.
16.
Veale
MF
,
Healey
G
,
Sparrow
RL
.
Longer storage of red blood cells is associated with increased in vitro erythrophagocytosis
.
Vox Sang
.
2014
;
106
(
3
):
219
-
226
.
17.
Bernardo
L
,
Amash
A
,
Marjoram
D
,
Lazarus
AH
.
Antibody-mediated immune suppression is improved when blends of anti-RBC monoclonal antibodies are used in mice
.
Blood
.
2016
;
128
(
8
):
1076
-
1080
.
18.
Samson
D
,
Mollison
PL
.
Effect on primary Rh immunization of delayed administration of anti-Rh
.
Immunology
.
1975
;
28
(
2
):
349
-
357
.
19.
Lubenko
A
,
Williams
M
,
Johnson
A
,
Pluck
J
,
Armstrong
D
,
MacLennan
S
.
Monitoring the clearance of fetal RhD-positive red cells in FMH following RhD immunoglobulin administration
.
Transfus Med
.
1999
;
9
(
4
):
331
-
335
.
20.
Nandakumar
KS
,
Collin
M
,
Olsén
A
, et al
.
Endoglycosidase treatment abrogates IgG arthritogenicity: importance of IgG glycosylation in arthritis
.
Eur J Immunol
.
2007
;
37
(
10
):
2973
-
2982
.
21.
Albert
H
,
Collin
M
,
Dudziak
D
,
Ravetch
JV
,
Nimmerjahn
F
.
In vivo enzymatic modulation of IgG glycosylation inhibits autoimmune disease in an IgG subclass-dependent manner
.
Proc Natl Acad Sci USA
.
2008
;
105
(
39
):
15005
-
15009
.
22.
Harvey
DJ
,
Merry
AH
,
Royle
L
,
Campbell
MP
,
Dwek
RA
,
Rudd
PM
.
Proposal for a standard system for drawing structural diagrams of N- and O-linked carbohydrates and related compounds
.
Proteomics
.
2009
;
9
(
15
):
3796
-
3801
.
Sign in via your Institution