Structural basis for HCMV Pentamer recognition by antibodies and neuropilin 2
Daniel Wrapp
1*
, Xiaohua Ye
2*
, Zhiqiang Ku
2
, Hang Su
2
, Harrison G. Jones
1
, Nianshuang Wang
1
, Akaash
K. Mishra
1
, Daniel C. Freed
3
, Fengsheng Li
3
, Aimin Tang
3
, Leike Li
2
, Dabbu Kumar Jaijyan
4
, Hua Zhu
4
,
Dai Wang
3
, Tong-Ming Fu
2
, Ningyan Zhang
2
, Zhiqiang An
2#
, Jason S. McLellan
1#
1
Department of Molecular Biosciences, The University of Texas at Austin, Austin, TX 78712, USA.
2
Texas Therapeutics Institute, Brown Foundation Institute of Molecular Medicine, University of Texas
Health Service Center at Houston, Houston, TX 77030, USA.
3
Merck Research Laboratories, Merck & Co., Inc., Kenilworth, NJ 07033, USA.
4
Department of Microbiology, Biochemistry and Molecular Genetics, Rutgers New Jersey Medical School,
Newark, NJ 07103, USA.
*
These authors contributed equally
#
Correspondence to zhiqiang.an@uth.tmc.edu (Z.A.) and jmclellan@austin.utexas.edu (J.S.M.)
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted March 25, 2021. ; https://doi.org/10.1101/2021.03.25.436804doi: bioRxiv preprint
ABSTRACT
Human cytomegalovirus (HCMV) encodes for multiple surface glycoproteins and glycoprotein
complexes
1,2
. One of these complexes, the HCMV Pentamer (gH, gL, UL128, UL130 and UL131),
mediates tropism to both epithelial and endothelial cells by interacting with the cell surface receptor
neuropilin 2 (NRP2)
3,4
. Despite the critical nature of this interaction, the molecular determinants that
govern NRP2 recognition remain unclear. Here we describe the cryo-EM structure of NRP2 bound to the
HCMV Pentamer. The high-affinity interaction between these proteins is calcium-dependent and differs
from the canonical C-terminal arginine (CendR) binding that NRP2 typically utilizes
5,6
. The interaction is
primarily mediated by NRP2 domains a2 and b2, which interact with UL128 and UL131. We also
determine the structures of four human-derived neutralizing antibodies in complex with the HCMV
Pentamer to define susceptible epitopes. The two most potent antibodies recognize a novel epitope yet
do not compete with NRP2 binding. Collectively, these findings provide a structural basis for HCMV
tropism and antibody-mediated neutralization, and serve as a guide for the development of HCMV
treatments and vaccines.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted March 25, 2021. ; https://doi.org/10.1101/2021.03.25.436804doi: bioRxiv preprint
MAIN TEXT 1
Human cytomegalovirus (HCMV) is a ubiquitous pathogen and congenital infection can cause 2
debilitating and permanent birth defects
7-9
. Despite the severity of these infections and the prevalence of 3
this pathogen, there are currently no FDA-approved vaccines and therapeutic options are limited
10-12
. 4
HCMV is an enveloped, double-stranded DNA virus of the family Herpesviridae
13
. The surface of the viral 5
membrane is decorated by several glycoprotein complexes that mediate viral entry and membrane 6
fusion
14-16
. One of these complexes is the HCMV Trimer, composed of glycoproteins gH, gL, and gO
2,17
. 7
The HCMV Trimer mediates tropism for fibroblasts by binding platelet derived growth factor receptor 8
alpha (PDGFRα)
18,19
. The HCMV Trimer is also capable of mediating infection of a broader variety of cell 9
types by interacting with transforming growth factor beta receptor 3 (TGFβR3)
4,20
. The other critical 10
tropism-determining complex is the HCMV Pentamer, which is composed of glycoproteins UL128, UL130, 11
UL131, and the same gH and gL proteins that comprise the bulk of the HCMV Trimer
1,3
. This elongated 12
heteropentamer mediates tropism for endothelial and epithelial cells by binding to neuropilin 2 (NRP2) 13
and triggering the viral fusion protein, gB, to facilitate viral entry into host cells
4,15,21-23
. 14
Neuropilins 1 and 2 are single-pass transmembrane proteins that are expressed on the surface of 15
neuronal, epithelial, and endothelial cells
24,25
. Under normal conditions, these proteins function as 16
receptors and co-receptors that engage in numerous physiological processes, including angiogenesis and 17
development of the nervous system
26,27
. NRP2 is composed of two N-terminal CUB domains (a1 and a2), 18
two F5/8 domains (b1 and b2), a MAM domain, a transmembrane domain, and a C-terminal PDZ domain 19
that is thought to mediate cytoplasmic signaling in response to extracellular stimuli
28,29
. Perhaps the most 20
thoroughly characterized of these stimuli is vascular endothelial growth factor (VEGF)
30
. The crystal 21
structure of these proteins in complex with one another has been determined, revealing that the b1 22
domain of NRP2 engages the C-terminal arginine of VEGF
6,31
. Since this initial characterization, the 23
NRP2 b1 domain has been shown to interact with other binding partners via the same mechanism
32
, 24
prompting the moniker “CendR” to refer to this exposed C-terminal arginine motif
5
. Although it has been 25
shown that soluble NRP2 is capable of inhibiting HCMV infection of epithelial cells
4
, the molecular 26
determinants that mediate this interaction remain unclear, and several additional Pentamer receptors 27
have been proposed
33,34
. The most potently neutralizing HCMV-directed antibodies are elicited against 28
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted March 25, 2021. ; https://doi.org/10.1101/2021.03.25.436804doi: bioRxiv preprint
the Pentamer, suggesting that it represents a susceptible target for the development of vaccines and 29
immunotherapeutics
35-37
. 30
To investigate NRP2 and mAb binding, we initiated structural and biophysical studies. Based on 31
previous crystallographic experiments that reported conserved calcium-coordinating loops in both the a1 32
and a2 domains of NRP2
38,39
, we measured the affinity of recombinantly expressed NRP2 a1a2b1b2 for 33
the soluble HCMV Pentamer ectodomain in both the presence and absence of calcium. We found that in 34
the presence of 2 mM EDTA, no association between NRP2 and Pentamer could be detected. However, 35
when the same experiment was performed in the presence of 2 mM CaCl
2
, the affinity of the interaction 36
was determined to be 2.2 nM (Supplementary Fig. 1a-b). It is possible that a failure to add additional 37
calcium is what necessitated the use of chemical cross-linkers during previous efforts to observe this 38
complex by negative-stain electron microscopy
4
. The addition of 2 mM CaCl
2
enabled us to form a stable 39
~230 kDa complex that was suitable for cryo-EM screening. The addition of 0.1% amphipol A8-35 helped 40
to prevent aggregation and allowed for the determination of a 4.0 Å resolution cryo-EM structure of the 41
HCMV Pentamer bound by human NRP2 (Fig. 1a, Supplementary Figs. 2 and 3). Performing focused 42
refinement on the NRP2-bound UL proteins yielded a 3.65 Å reconstruction that aided in model building 43
and refinement. 44
These reconstructions revealed an extensive binding interface, with contacts formed by NRP2 45
domains a1, a2 and b2 (Fig. 1). Notably, the calcium-coordinating loop of domain a2 (residues 251−258) 46
forms a sizable portion of this binding interface, likely providing an explanation as to why high-affinity 47
NRP2 binding could only be observed after the addition of 2 mM CaCl
2
(Fig. 1a-b). Additional contacts 48
are formed between the C-terminal beta strands of ULs 130 and 131 and a loop formed by residues 453-49
461 of the b2 domain of NRP2 (Fig. 1c). This mode of NRP2 binding differs from the canonical CendR 50
motif binding that has been described previously for other NRP2-binding partners
40,41
. The CendR binding 51
mechanism involves the engagement of a C-terminal arginine residue by the b1 domain, whereas 52
Pentamer is exclusively bound by the a1, a2, and b2 domains. Furthermore, none of the three UL 53
proteins contain a positively charged C-terminal arginine that makes up the CendR motif. As expected, 54
the majority of the binding interface from the Pentamer is composed of the tropism-determining UL 55
proteins, particularly UL128 and UL131
1,3
, which respectively contribute 437.5 Å
2
and 208.4 Å
2
of buried 56
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted March 25, 2021. ; https://doi.org/10.1101/2021.03.25.436804doi: bioRxiv preprint
surface area to the interface. Whereas the NRP2 a2, b1, and b2 domains are clustered tightly together at 57
the head of the Pentamer, the N-terminal a1 domain is tethered via a flexible linker that allows it to bind 58
near the middle of the Pentamer, where the C-terminus of UL128 associates with gL. The local resolution 59
for this portion of the reconstruction was relatively poor compared to the rest of the complex, suggesting 60
either conformational flexibility in this region or a loose association of a1. To test the importance of the a1 61
domain, we expressed NRP2 with a 144-residue N-terminal truncation and observed that even in the 62
absence of this flexibly tethered a1 domain, NRP2 a2b1b2 was capable of binding to the HCMV 63
Pentamer with 7.9 nM affinity (Supplementary Fig. 1c), supporting our structural observations that the 64
critical determinants of Pentamer binding are contained within NRP2 domains a2b1b2. Intriguingly, our 65
cryo-EM data processing also revealed that a second, more poorly resolved copy of NRP2 could be 66
observed binding near the C-terminal arginine of gL via the b1 domain (Supplementary Figs. 2 and 4). 67
Although this second NRP2 appears to exhibit the canonical CendR binding, it could only be observed in 68
~40% of particles. Furthermore, its binding to the gL protein rather than the tropism-determining UL 69
proteins suggests that this second copy of NRP2 is likely an artifact of the high concentrations of NRP2 70
that were used to form a stable complex. Overall, the conformation of the receptor-bound Pentamer 71
ectodomain does not drastically differ from that of the unbound Pentamer
42
(Supplementary Fig. 5), 72
suggesting that rather than undergoing substantial conformational rearrangements, this complex acts as 73
a tether to connect HCMV virions to the surface of epithelial and endothelial cells until the viral fusogen 74
gB fuses the viral and cellular membranes. 75
Previous efforts to characterize the humoral immune response to asymptomatic HCMV infection 76
yielded an extensive panel of neutralizing antibodies directed against gB, the HCMV Trimer, and the 77
HCMV Pentamer
35,43,44
. To learn more about the mechanisms of neutralization of high-affinity, Pentamer-78
directed antibodies, we determined cryo-EM structures of four naturally elicited human antibodies in 79
complex with the Pentamer (Fig. 2, Supplementary Figs. 3, 6, 7, and 8). The flexibility and elongated 80
shape of the Pentamer necessitated focused refinements of the Fabs along with the domains making up 81
their respective epitopes. Model building was facilitated by high-resolution crystal structures of unbound 82
Fabs (1-103: 1.9 Å, 1-32: 2.1 Å, 2-18: 2.8 Å, 2-25: 2.5 Å), which were then used as reference restraints 83
and lightly refined as a part of the complex (Supplementary Tables 1 and 2). Three of these antibodies 84
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted March 25, 2021. ; https://doi.org/10.1101/2021.03.25.436804doi: bioRxiv preprint