Article

ISSN 0006-2979, Biochemistry (Moscow), 2024, Vol. 89, No. 7, pp. 1260-1272 © The Author(s) 2024. This article is an open access publication.

1260

Structural Basis for Evasion of New SARS-CoV-2 Variants

from the Potent Virus-Neutralizing Nanobody Targeting

the S-Protein Receptor-Binding Domain

Nikolai N. Sluchanko

1,2,a

*, Dmitry V. Shcheblyakov

2,b

*, Larisa A. Varfolomeeva

Irina A. Favorskaya

, Inna V. Dolzhikova

, Anastasia I. Korobkova

Irina A. Alekseeva

, Ilias B. Esmagambetov

, Artem A. Derkaev

Vladimir V. Prokofiev

, Ilya D. Zorkov

, Denis Y. Logunov

, Alexander L. Gintsburg

Vladimir O. Popov

, and Konstantin M. Boyko

1,2,c

Bach Institute of Biochemistry, Federal Research Centre “Fundamentals of Biotechnology”,

Russian Academy of Sciences, 119071 Moscow, Russia

National Research Center for Epidemiology and Microbiology named after Honorary Academician N.F. Gamaleya,

Ministry of Health of the Russian Federation, 123098 Moscow, Russia

e-mail: nikolai.sluchanko@mail.ru

e-mail: sdmitriyv@mail.ru

e-mail: kmb@inbi.ras.ru

Received May 16, 2024

Revised June 4, 2024

Accepted June 6, 2024

Abstract— COVID-19 has caused millions of deaths and many times more infections worldwide, emphasizing the

unpreparedness of the global health system in the face of new infections and the key role for vaccines and thera-

peutics, including virus-neutralizing antibodies, in prevention and containment of the disease. Continuous evolu-

tion of the SARS-CoV-2 coronavirus has been causing its new variants to evade the action of the immune system,

which highlighted the importance of detailed knowledge of the epitopes of already selected potent virus-neutral-

izing antibodies. A single-chain antibody (“nanobody”) targeting the SARS-CoV-2 receptor-binding domain (RBD),

clone P2C5, had exhibited robust virus-neutralizing activity against all SARS-CoV-2 variants and, being a major

component of the anti-COVID-19 formulation “GamCoviMab”, had successfully passed PhaseI of clinical trials.

However, after the emergence of the Delta and XBB variants, a decrease in the neutralizing activity of this nano-

body was observed. Here we report on the successful crystal structure determination of the RBD:P2C5 complex

at 3.1Å, which revealed the intricate protein–protein interface, sterically occluding full ACE2 receptor binding by

the P2C5-neutralized RBD. Moreover, the structure revealed the developed RBD:P2C5 interface centered around

residues Leu452 and Phe490, thereby explaining the evasion of the Delta or Omicron XBB, but not Omicron

B.1.1.529 variant, as a result of the single L452R or F490S mutations, respectively, from the action of P2C5.

The structure obtained is expected to foster nanobody engineering in order to rescue neutralization activity and

will facilitate epitope mapping for other neutralizing nanobodies by competition assays.

DOI: 10.1134/S0006297924070083

Keywords: crystal structure, protein–protein interaction, neutralizing antibodies

* To whom correspondence should be addressed.

INTRODUCTION

Infectious disease outbreaks, whether pandemic

or seasonal, impose significant economic and social

burdens. The therapeutic use of virus-neutralizing an-

tibodies has emerged as a promising strategy against

certain infectious diseases [1-5]. However, the continu-

al evolution of viral strains during ongoing epidemics

can undermine the effectiveness of existing treatments.

Mutations in emerging variants can diminish the po-

tency of established antibody therapies [6], underscor-

ing the need for detailed structural insights into virus-

neutralizing antibody epitopes to bolster therapeutic

design through protein engineering methods [7, 8].

STRUCTURE OF THE SARS-CoV-2 RBD:P2C5 NANOBODY COMPLEX 1261

BIOCHEMISTRY (Moscow) Vol. 89 No. 7 2024

Abbreviations: ACE2,angiotensin-converting enzyme2; CDR,complementarity-determining region; COVID-19,Coronavirus

disease 2019; RBD,receptor-binding domain; RMSD,root-mean-square deviation of atomic positions; SARS-CoV-2,severe

acute respiratory syndrome-related coronavirus2.

Amidst the COVID-19 (Coronavirus disease 2019)

pandemic, which has claimed over 7 million lives glob-

ally from 2019 to April 2024 (https://covid19.who.int),

researchers worldwide have mobilized efforts to de-

velop monoclonal antibodies with potent virus-neu-

tralizing capabilities. These efforts have predominant-

ly targeted the SARS-CoV-2 (severe acute respiratory

syndrome-related coronavirus 2) surface glycoprotein

(S,or Spike protein), aiming to disrupt its interaction

with the ACE-2 cellular receptor and thwart viral entry

into human cells [9-13].

In the National Research Center for Epidemiol-

ogy and Microbiology, named after Honorary Aca-

demician N. F. Gamaleya of the Ministry of Health of

the Russian Federation, a drug called GamCoviMab

based on a mix of two neutralizing antibodies, in-

cluding the P2C5 clone, has been developed (https://

classic.clinicaltrials.gov/ct2/show/NCT05729360). This

drug demonstrated efficacy in a model of lethal ACE-2

transgenic mouse infection with various SARS-CoV-2

variants, including the Omicron BA.1, BA.2, and BA.5

variants. Evaluation of the drug’s in vitro virus-neu-

tralizing activity against these Omicron variants re-

vealed a minimum effective concentration of less than

1.2 μg/ml [14, 15]. P2C5-containing preparation has

successfully passed PhaseI of clinical trials. However,

after the emergence and wide distribution of the Delta

coronavirus variant harboring an L452R mutation and

Omicron XBB.1 variant harboring an F490S mutation

in the receptor-binding domain (RBD), a decrease in

the neutralizing activity of this clone was found.

Novel SARS-CoV-2 variants, such as Delta and

Omicron lineages, have raised substantial concerns

due to their potential to evade immune responses and

resist current therapeutic interventions based on pre-

viously very potent neutralizing antibodies [16-19].

Inresponse to these evolving challenges, research into

the mechanism of neutralization and antibody evasion

by novel variants with mutated RBD of the S protein

has garnered significant attention.

This study reports on the successful crystal struc-

ture determination and analysis of the complex be-

tween the virus-neutralizing single-chain antibody

P2C5 and either glycosylated or deglycosylated forms

of the Wuhan SARS-CoV-2 S RBD.

MATERIALS AND METHODS

Cell lines and viruses. CHO-S cell line was

purchased from Thermo Fisher Scientific (USA),

cat. no. R80007. VeroE6 cells (ATCC CRL 1586) were

obtained from the Russian Collection of Cell Cultures

of Vertebrates (CCCV). Cells were cultured at 37°C and

5% CO

in Dulbecco’s Modified Eagle Medium (DMEM;

Gibco, Switzerland), supplemented with 10% (v/v) fetal

bovine serum (Hyclone/Cytiva, USA).

SARS-CoV-2 strains B.1.1.1 (Wuhan D614G, hCoV-19/

Russia/ Moscow_PMVL-1/2020), B.1.617.2 (Delta, hCoV-19/

Russia/SPE-RII-32758S/2021), B.1.1.529.1 (Omicron BA.1,

hCoV-19/Russia/MOW-Moscow_PMVL-O16/2021), Omi-

cron XBB.1.17.2 (hCoV-19/Russia/SPE-RII-4422S/2022)

were isolated from nasopharyngeal swabs.

The bacterial strain Escherichia Coli Rosetta

(DE3) was purchased from Merck-Millipore (USA), cat.

no.70954.

Recombinant proteins expression and puri-

fication. Plasmid DNA containing P2C5 coding se-

quence was transformed into E. coli Rosetta DE3

(Merck- Millipore). The transformed cells were cul-

tured in 2xYT medium supplemented with ampicillin

(100μg/ml) at37°C at 210rpm to OD

600

0.5-0.8. Then IPTG

(Anatrace, USA) was added (0.1mM) and the culture

was grown at 30°C overnight. Recombinant P2C5 with

a C-terminal 6×His-tag was isolated from the E. coli

periplasm by cold osmotic shock and purified by met-

al-ion affinity chromatography, as described in previ-

ous work [15, 20]. Also, P2C5 VHH was transformed to

a P2C5-Fc format by fusing P2C5 nucleotide sequence

to the human IgG1 Fc-fragment sequence. The obtained

construction was cloned into the mammalian expres-

sion vector pCEP4 (Thermo Fisher Scientific).

The Spike protein receptor-binding domain (RBD)

sequences of different variants of SARS-CoV-2 (Wuhan-

Hu-1, Gamma, Delta, and Omicron XBB) and also

N-terminally truncated RBD of Wuhan-Hu-1 variant

sequence (coding amino acid residues 333-541) were

cloned into the pCEP4 vector. The information con-

cerning RBD sequences and mutations are summarized

in Fig.4b. The RBD expression constructs contained

C-terminal polyhistidine tags for further purification

steps.

For recombinant protein expression, CHO-S cells

(Thermo Fisher Scientific) were transfected with the

obtained pCEP4 vectors using the CHOgro™ Expression

System (Mirus Bio, USA) according to the manufac-

turer’s instructions. Polyhistidine-tagged recombinant

proteins were purified from cell culture superna-

tants using metal-ion affinity chromatography resin

Sepharose 6 Fast Flow, AKTA Start system and HisTrap

HP column (Cytiva). The second purification step was

performed using size exclusion chromatography res-

in Superdex200 (Cytiva). P2C5-Fc was purified using

HiTrap ProteinA (Cytiva). The purity of recombinant

SLUCHANKO et al.1262

BIOCHEMISTRY (Moscow) Vol. 89 No. 7 2024

proteins was assessed by Coomassie Blue staining of

SDS-PAGE gels.

Peptide-N-Glycosidase F (PNGase F) for optional

RBD deglycosylation was produced in the recombinant

form in previous work [20].

ELISA. 96-well immunoplates (Greiner, Austria)

were coated with antigens (1μg/ml) overnight at 4°C.

Wells were blocked with 5% non-fat milk (PanReac

AppliChem, Spain) in phosphate-buffered saline with

0.05% Tween-20 (TPBS). Serial three-fold dilutions of

P2C5 (10 µg/ml to 4.5 ng/ml) were added to the wells.

Then horseradish peroxidase (HRP)-conjugated rab-

bit anti-Myc tag antibodies (ab1326, Abcam, UK) were

added. EC50 values were calculated using GraphPad

Prism 7 software. In this study, we used the following

antigens: receptor-binding domains of spike proteins

of Wuhan, Gamma, Delta and XBB.1 virus variants,

S1 subdomain of spike protein of Omicron B.1.1.529

variant (Sino Biological) (RBD mutations in this variant

are the same as in Omicron BA.1).

Virus neutralization assay. Neutralization assay

with live SARS-CoV-2 was performed as previously

described [15]. Serial two-fold dilutions of P2C5 an-

tibodies at concentrations ranging from 20mg/ml to

1.2 ng/ml were mixed with 100 TCID

of SARS-CoV-2

virus (Omicron XBB.1.17.2 or B.1.1.1). The mixture was

incubated for 1 h at 37°C and added to a monolayer

of Vero E6 cells. After incubation (96-120 h) at 37°C

in 5% CO

, the neutralizing activity of the antibody

was assessed visually by the ability to inhibit the cy-

topathic effect of the virus (CPE). The minimum neu-

tralizing concentration of P2C5 was determined as the

minimum antibody concentration at which complete

inhibition of the cytopathic effect was observed. All

experiments were performed with three replicates.

In vivo protection studies in mice. Hemizygous

angiotensin-converting enzyme 2 (ACE2)-transgenic

mice (Tg (K18-ACE) 2Prlmn) 10-12 weeks old were used

to evaluate the protective capacity of P2C5-Fc antibod-

ies in an invivo model of SARS-CoV-2 infection. Mice

were kept in individually ventilated cages (ISOcage P

System from Tecniplast) and had free access to food

and water. Procedures with animals were conduct-

ed in a biosafety level3 (BSL3) laboratory. All animal

experiments were carried out in strict accordance

with the recommendations of the Russian Federation

(GOST R 53434-2009; Principles of Good Laboratory

Practice).

The mice were randomly divided into groups of

5-8 animals each. Infection of animals was performed

via intranasal administration of 1×10

TCID

of SARS-

CoV-2 virus. In the first experiment, two groups of mice

(5 animals per group) were intraperitoneally injected

with 5 mg/kg of P2C5-Fc 1 and 6 hours after infection

by B.1.1.1 virus variant. The control group (n = 5) was

treated with PBS. Animals were monitored for survival

rate and weight change for 20 days post infection. Mice

with 20% body weight loss were humanely euthanized.

In the following experiment, to assess the protective

capacity of P2C5-Fc antibodies against Delta (B.1.617.2)

SARS-CoV-2 variant, a group of mice (n = 5) was admin-

istered intraperitoneally with 5 mg/kg of P2C5-Fc 1 h af-

ter B.1.617.2 challenge. Control animals were injected

with PBS. All procedures in this study were performed

as described above.

Inoculation with SARS-CoV-2 of Omicron B.1.1.529

variant was not lethal in hACE2 mice. To evaluate

the therapeutic efficacy of P2C5-Fc against Omicron

B.1.1.529, mice (n=8) were injected intraperitoneally

with 10 mg/kg of P2C5-Fc 1 h after infection. A control

group of 8 animals received PBS. Viral titer (TCID

)

was measured in lung tissue 4 days after infection.

Lung homogenates were prepared using an MPBio

homogenizer. Serial ten-fold dilutions of homogenates

were titrated in a monolayer of Vero E6 cells to deter-

mine the titer of the infectious virus. The cytopathic

effect of the virus was assessed visually after 96 h.

TCID

was calculated by the Reed and Muench meth-

od. Titers below the limit of detection were taken

as1.5 log

 TCID

/ml.

Preparation and crystallization of the RBD-P2C5

VHH complex. The detailed procedure of the prepara-

tion of the RBD-P2C5 VHH complex was described ear-

lier [20]. Briefly, the purified proteins were pre-mixed

at different ratios and analyzed by size-exclusion chro-

matography on a Superdex 200 Increase 5/150 column

(Cytiva) to empirically determine the ratio correspond-

ing to a slight excess of the nanobody. Then, milligram

quantities of the proteins were mixed at this chosen

volume ratio, incubated, and the heterodimeric com-

plex was separated from the nanobody excess using

Superdex 200 Increase 10/300 column (Cytiva). For RBD

deglycosylation, the sample loaded on preparative SEC

was first treated with PNGase F for 2 h at room tem-

perature and then overnight in the fridge. The complex

fractions were pooled and concentrated before crys-

tal screening using Hampton Research crystallization

kits (USA) and an Oryx4 crystallization robot (Doug-

las Instruments, UK). The best crystals of RBD

319-541

P2C5-VHH and enzymatically deglycosylated RBD

333-541

P2C5-VHH complexes were obtained at the following

conditions: 0.2 M Ammonium sulfate, 0.1 M Bis-Tris

pH 6.5, 25% PEG 3350 and 0.15 M DL-Malic acid pH 7.0,

0.1 M Imidazole pH 7.0, 22% v/v Polyethylene glycol

monomethyl ether 550, respectively. 20% ethylene gly-

col was used as a cryoprotectant.

Crystal structure determination. The structure

of the N-terminally truncated and PNGase F-treated

RBD (residues 333-541) complexed with P2C5 VHH

was solved by molecular replacement using MolRep

[21] and RBD (PDB ID 7A91) as a search model, which

initially resulted in four RBD copies correctly found

STRUCTURE OF THE SARS-CoV-2 RBD:P2C5 NANOBODY COMPLEX 1263

BIOCHEMISTRY (Moscow) Vol. 89 No. 7 2024

and several misplaced copies requiring manual de-

letion based on the electron density maps revealed.

Further molecular replacement using the partial solu-

tion composed of four RBD copies allowed to find

eight copies of the nanobody modeled by AlphaFold2

[22] based on the amino acid sequence of P2C5 VHH.

Of these eight, only one copy of the nanobody was

placed close to an RBD and matched the density. This

tentative heterodimeric complex was used as a new

search model for molecular replacement procedure in

MolRep [21], which allowed us to find four copies of

the heterodimeric complex between RBD and the na-

nobody. This partial solution revealed a strong differ-

ence density suggesting the correctness of the partial

solution and the presence of more protein molecules

in the asymmetric unit. To find those, the heterodimer

was again used as a search model in MolRep while

using the previously found partial solution composed

of the four copies of the complex. Finally, this result-

ed in placing 5 heterodimeric complexes in the asym-

metric unit, enabling the refinement of the complete

structure, which was done iteratively using Refmac5

[23] and model building in Coot [24]. The refinement

involved backbone H-bond restraints, NCS restraints

and use of TLS. At final steps of refinement, the use

of the ProSMART option in the Refmac5 and reference

structures of RBD (PDB ID 7FAT [25]) and nanobody

(PDB ID 3STB [26]) helped improve geometry and the

R-factors. The refined model of the heterodimeric com-

plex between RBD

333-541

and P2C5-VHH was used to

also solve the similar structure between P2C5-VHH and

untruncated RBD

319-541

not treated with PNGaseF for

deglycosylation.

RESULTS AND DISCUSSION

A single-domain camelid antibody P2C5 with po-

tent neutralizing activity against a range of SARS-CoV-2

virus variants has previously been identified [15].

Thecapacity of P2C5-VHH to recognize the S protein of

different SARS-CoV-2 variants was evaluated by ELISA

(Fig. 1a). We found that P2C5 exhibited strong binding

activity toward the S protein of Wuhan, Gamma, and

Omicron B.1.1.529 variants. However, binding to the

Sprotein of Delta and XBB.1 decreased, likely due to

the mutations affecting the epitope region. Interest-

ingly, fusing P2C5 with Fc resulted in nearly 10 times

more efficient RBD recognition (Fig.1b).

The ability of P2C5 to neutralize live SARS-CoV-2

virus was investigated by microneutralization assay.

The virus was mixed with antibodies and then add-

ed to Vero E6 cells. The absence of cytopathic effect

in the presence of P2C5 was the key evidence of the

antibody’s neutralization activity. A combination of

current data on virus neutralization by the P2C5 anti-

body with the previously published data [15] is found

in Table 1. We observed that P2C5 completely inhibit-

ed the cytopathic effect of SARS-CoV-2 Wuhan D614G

(B.1.1.1), Gamma (B.1.1.28/P.1), and Omicron B.1.1.529

variants at concentrations of 24-48nM. The inhibition

of the cytopathic effect of SARS-CoV-2 Delta (B.1.617.2)

and Omicron XBB.1.17.2 variants was not observed

over a wide range of P2C5 concentrations (to 1500nM).

Theinvitro neutralization activity data are consistent

with binding ability data obtained by ELISA.

The protective capacity of P2C5 against SARS-

CoV-2 infection in vivo was assessed using hACE2

Fig. 1. Recognition of S protein of SARS-CoV-2 variants by P2C5 single-domain antibody(a) or P2C5-Fc(b). The binding of P2C5

to immobilized antigens (RBD of S protein for Wuhan, Gamma, Delta, and Omicron XBB.1 variants and S1 subdomain of S pro-

tein for Omicron B.1.1.529 variant) was detected by ELISA. P2C5 bound S protein of Wuhan, Gamma, and Omicron B.1.1.529

variants with half-maximal concentration (EC50) 1.7, 7.6, 2.9nM, respectively.

SLUCHANKO et al.1264

BIOCHEMISTRY (Moscow) Vol. 89 No. 7 2024

Table 1. Neutralizing activity of P2C5 antibody against live SARS-CoV-2 variants

SARS-CoV-2 virus variant

B.1.1.1

Gamma

(B.1.1.28/P.1)*

Delta (B.1.617.2)*

Omicron

B.1.1.529*

Omicron

XBB.1.17.2

Minimum inhibitory

concentration

of P2C5 VHH, nM

24.04 48.08 >1500 24.04 >1500

* Previously published data from [15].

transgenic mice. For these studies, P2C5 was fused to

the human IgG1 Fc-fragment for half-life prolonga-

tion. The P2C5-Fc protein was produced in CHO-S cells

and purified. We assessed P2C5-Fc therapeutic efficacy

against infection caused by three different SARS-CoV-2

variants: Wuhan D614G (B.1.1.1), Delta, and Omicron

B.1.1.529.

First, we evaluated the protective effect of P2C5-Fc

after B.1.1.1 (Wuhan D614G) and B.1.617.2 (Delta) le-

thal challenge. Mice were infected intranasally with

1 × 10

TCID

of SARS-CoV-2 and 1 h or 6 h later re-

ceived 5 mg/kg P2C5-Fc antibodies intraperitoneally.

As shown in Fig.2a, all mice treated with P2C5-Fc sur-

vived after B.1.1.1 infection (p = 0.0004, log-rank test),

furthermore, no weight loss was observed. After infec-

tion with the Delta variant of the virus, surprisingly,

40% of antibody-treated animals survived (data not

significant, p = 0.33, log-rank test), despite the fact that

single-domain P2C5 antibody did not recognize the S

protein of this variant in ELISA (Fig.2b). A probable

explanation for the revealed potential protection is the

capacity of the dimeric Fc-fused form of P2C5 to weak-

ly bind the S protein of Delta (Fig.1b).

The SARS-CoV-2 Omicron B.1.1.529 variant con-

tains a large number of mutations resulting in non-le-

thal virulence in hACE2 mice. As all control group

mice (treated with PBS after infection) survived and

showed no marked weight loss, we assessed the virus

load in the lungs of the animals on day 4 after infec-

tion (Fig.2c). Administration of P2C5-Fc 1 h after chal-

lenge resulted in a significant reduction of SARS-CoV-2

titer in the lungs (p = 0.007, Mann–Whitney test), and

the virus content in the antibody-treated group was

below the limit of detection (<1.5 log

TCID

/ml). This

result highlights the efficacy of P2C5-Fc treatment

invivo also against the Omicron B.1.1.529 variant of

the SARS-CoV-2 virus.

The differential activity of P2C5 on SARS-CoV-2

variants clearly required mechanistic explanation.

Tothis end, we aimed at crystal structure determina-

tion of the RBD:P2C5 complex and subjected the pu-

rified complex to extensive crystallization screening

[20]. Although this complex readily crystallized un-

der various conditions, it proved difficult to find dif-

fracting crystals and therefore we attempted different

sample modifications. In particular, we subjected to

crystallization untreated RBD or RBD deglycosylated

by recombinant PNGaseF [20], in the hope that degly-

cosylation would reduce the heterogeneity associated

with RBD. In addition, RBD was truncated from the

N-terminus until residue 333 so that several upstream

glycosylated residues were omitted. Wide screening of

many crystals using synchrotron X-ray radiation was

crucial to finding diffracting crystals for both unmod-

ified RBD

319-541

:P2C5 and enzymatically deglycosylat-

ed RBD

333-541

:P2C5 complexes (Table2). Given that the

diffraction quality of the two crystals was comparable

(3.1 or 3.7 Å), we could finally suggest that deglycosyla-

tion on its own was dispensable for obtaining crystals

of sufficient quality, which disproved our initial con-

siderations.

The crystal structure was first solved for the min-

imally glycosylated RBD

333-541

:P2C5 complex, and the

solution was then used to also determine the struc-

ture of the RBD

319-541

:P2C5 complex, which turned to

be nearly identical in terms of the relative positions of

RBD and P2C5 within the heterodimer [Cα RMSD (root-

mean-square deviation of atomic positions) did not

exceed 0.5 Å]. Therefore, for further analysis we chose

the RBD

319-541

:P2C5 structure as having the higher res-

olution (Fig.3a), which was sufficient to confidently

trace most of the core RBD and P2C5 residues into the

electron density maps (Fig.3b).

The asymmetric unit of the crystal contained five

copies of the RBD

319-541

:P2C5 heterodimer totaling 1575

residues (Fig.3a), and these copies were nearly iden-

tical (Cα RMSD did not exceed 0.5 Å) (Fig.3c). Inter-

estingly, the final refined crystal structure of the het-

erodimer was notably different from the closest model

of the complex predicted by AlphaFold2 (at the end of

April 2024), providing Cα RMSD of only 4.6 Å upon su-

perposition of 311 atoms of the complex. Fig.3d shows

the overlay of these two structures aligned by RBD.

Itis clear that, while AlphaFold2 rather accurately pre-

dicted the RBD structure (Cα RMSD between the crystal

structure and the model did not exceed 0.5 Å), the mod-

el of its complex with the nanobody was inadequate.

Intriguingly, the very recently released AlphaFold3

STRUCTURE OF THE SARS-CoV-2 RBD:P2C5 NANOBODY COMPLEX 1265

BIOCHEMISTRY (Moscow) Vol. 89 No. 7 2024

Fig. 2. Therapeutic efficacy of P2C5-Fc against SARS-CoV-2 infection in hACE mice. a-b)Survival rates and mean body weights of

mice (n=5 per group) that were systemically treated (i.p.) with P2C5-Fc 1h or 6h after intranasal inoculation of 1 × 10

TCID

of Wuhan D614G (B.1.1.1) or Delta (B.1.617.2) SARS-CoV-2 variants. The control group was treated with PBS 1h post infec-

tion. c)Lung viral titer on day 4 post infection with Omicron B.1.1.529 (1 × 10

TCID

intranasally) in mice injected(i.p.)

with 10mg/kg P2C5-Fc or vehicle (PBS) 1h after challenge (n=8 mice per group). Viral titers in the P2C5-Fc group were below

the limit of detection of the assay and are shown as 1.5log

TCID

/ml. **p=0.007.

neural network, upgraded for predicting complexes

[27], predicted a completely different P2C5 epitope on

the opposite side of RBD with respect to the X-ray struc-

ture and the AlphaFold2 model (Fig.3d). These rather

poor predictions by superior insilico algorithms sup-

port the notion that experimental structural biology

SLUCHANKO et al.1266

BIOCHEMISTRY (Moscow) Vol. 89 No. 7 2024

Table 2. X-ray data collection and processing

Crystallization condition

RBD

333-541

complexed

with P2C5 nanobody*

RBD

319-541

complexed

with P2C5 nanobody

0.15M DL-Malic acid pH7.0,

0.1M Imidazole pH7.0,

22% PEG MME 550

0.2M (NH

)

, 0.1M Bis-Tris,

pH6.5, 25% PEG 3350

Data collection

Diffraction source BL17UM, SSRF BL17UM, SSRF

Detector Eiger2 X 16M Eiger2 X 16M

Wavelength (Å) 0.98 0.98

Temperature (K) 100 100

Space group I2

a, b, c (Å) 126.64, 194.24, 264.37 72.48, 177.46, 209.86

Resolution range (Å) 80.00-3.70 (3.80-3.70)** 48.42-3.10 (3.20-3.10)

Completeness (%) 99.1 (99.6) 98.9 (98.6)

Redundancy 5.0 5.2

〈I/σ(I)〉 6.6 (0.6) 8.0 (1.5)

meas

(%) 22.8 (316.2) 19.6 (149.5)

1/2

(%) 99.6 (32.1) 99.3 (45.0)

Refinement

cryst

(%) 20.7 22.2

free

(%) 26.9 28.3

ML position error, Å 0.57 0.41

No. of non-H atoms

Protein 12,286 12,255

Other 112 220

Water 0 0

R.m.s. deviations

Bonds (Å) 0.02 0.02

Angles (°) 3.34 3.01

Ramachandran outliers (%) 3.4 3.7

Average B factors (Å

)

Protein 195.6 86.4

Other 228.0 145.0

Water – –

PDB code 8ZES 8ZER

* Treated with PNGase F for RBD deglycosylation.

** Values for the outer shell are given in parentheses.

STRUCTURE OF THE SARS-CoV-2 RBD:P2C5 NANOBODY COMPLEX 1267

BIOCHEMISTRY (Moscow) Vol. 89 No. 7 2024

Fig. 3. Structural basis for the virus-neutralizing activity of the P2C5 nanobody. a)Overall view of the asymmetric unit (ASU)

of the RBD:P2C5 crystal showing five heterodimers and glycans linked to residue Asn343 of RBD. b)Exemplary fragments of

the electron density map contoured at 1.2σ, interpreted with the final refined protein model shown as sticks. c)Structural

superposition of the five heterodimer complexes found in the ASU. Complementarity-determining region (CDR) loops 1-3 are

highlighted. d)Superposition of the final refined crystal structure of the complex and its best models predicted by AlphaFold2

(black cartoon) or AlphaFold3 (red cartoon) showing inadequacy of the insilico models. Cα RMSD of the alignment over the full

complex was equal to as large as 4.6Å for the AlphaFold2 model, whereas AlphaFold3 predicted a completely different P2C5

epitope. e)Overlay of the RBD:P2C5 crystal structure with the ACE2:Spike complex structure determined by cryoelectron mi-

croscopy (PDBID 8WTJ [17]) explaining the neutralization activity of P2C5 nanobody via steric interference with the RBD:ACE2

interaction. One Spike monomer is colored by gradient from N (blue) to C (red) terminus. Red and magenta spheres indicate

thelocation of the L452R and F490S mutations, respectively.

SLUCHANKO et al.1268

BIOCHEMISTRY (Moscow) Vol. 89 No. 7 2024

approaches are indispensable for studying antibody

complexes in particular.

In the crystallographic heterodimer, the comple-

mentarity-determining regions (CDRs) CDR3 (residues

96-110) and CDR2 (residues 50-57) of P2C5 nanobody

contributed to the RBD recognition the most, whereas

CDR1 (residues 26-33) did not partake in the interac-

tion directly (Fig. 3c). The relative position and orien-

tation of RBD and P2C5 in the crystallographic het-

erodimer has allowed us to analyze the corresponding

Fig. 4. Structural basis for evasion of the Delta and Omicron XBB.1 variants of SARS-CoV-2 from the action of the neutralizing

P2C5 nanobody. a)Crystallographic RBD:P2C5 heterodimer showing the location of key interface contacts. Polar contacts are

indicated by dashed lines, key mutated residues are marked by bold font. b)S protein RBD sequence and nonsynonymous mu-

tations in RBD of SARS-CoV-2 variants. Note that RBD of Omicron B.1.1.529 has coinciding mutations as RBD of Omicron BA.1.

Key mutations described in the manuscript are highlighted. c)Closeup view of the modeled interface between P2C5 (cyan)

andhypothetical RBD containing simultaneous substitutions L452R (Delta) and F490S (Omicron XBB.1).

STRUCTURE OF THE SARS-CoV-2 RBD:P2C5 NANOBODY COMPLEX 1269

BIOCHEMISTRY (Moscow) Vol. 89 No. 7 2024

interface in context of the recently published cryo-EM

structure [17] of the XBB Spike protein complexed with

the ACE2 receptor (Fig.3e). This comparison indicated

that P2C5 recognizes an epitope on the outer surface of

the RBD in a conformation which causes appreciable

steric clashes with the RBD-bound ACE2, at least partly

explaining the mechanism of neutralization. Interest-

ingly, the P2C5 epitope is located on the other side of

the RBD surface compared with the location of Leu455

and Phe456 residues, and therefore the activity of this

antibody is not expected to be affected by the recently

described, adjacent residue-flipping mutations L455F

and F456L, which synergistically knock out many an-

tibodies [6, 17].

Furthermore, the developed interface between

P2C5 and RBD covering ~940 Å

area, as calculated

using PISA [28], revealed many polar and hydropho-

bic contacts apparently stabilizing the interaction. For

example, Arg357 and Glu471 of the RBD formed salt

bridges with Glu104 and Arg52 of P2C5, respectively,

Tyr449 of the RBD formed π–cation interaction with

the side chain of Arg45 of P2C5, and also a range of

intersubunit H-bonds were observed (Fig.4a). Inter-

estingly, the interface featured a hydrophobic core

involving clustered aromatic residues from both RBD

(e.g., Phe490, Tyr351) and P2C5 (e.g., Tyr37, Phe47,

Trp111, Trp98) and also a few aliphatic hydropho-

bic residues among which Leu452 of RBD is the most

prominent. Intriguingly, this residue is known to be-

come mutated in the Delta variant (Fig.4b) and there-

fore its location in the center of the interface with

P2C5 nanobody suggested immediate consequences of

the Leu452 mutation on the stability of the complex.

Indeed, while the second Delta mutation, T478K, is lo-

cated on the outer surface of RBD and cannot direct-

ly affect the interaction with P2C5, L452R is the mu-

tation which presumably leads to the placement of

the bulky, positively charged Arg side chain opposite

to the long positively charged side chain of Lys96 of

P2C5, unavoidably causing electrostatic repulsion and

severe steric clashes especially given that the side

chain of Lys96 is firmly sandwiched between the side

chains of Trp98 and Trp111 of P2C5 (Fig.4c). This ob-

viously unfavorable configuration nicely explains the

evasion of the Delta variant from the action of P2C5

nanobody.

We also noticed that two heavily mutated Omicron

variants, B.1.1.529 and XBB.1 (Fig.4b), among which

only the former is efficiently recognized by P2C5 na-

nobody, share the E484A mutation located in the

RBD:P2C5 interface (Fig.4a). While Glu484 could be in-

volved in making a polar contact with the side chain of

Thr60 of P2C5, its substitution with Ala can break this

polar contact and potentially destabilize the RBD:P2C5

binding. Nevertheless, this effect seems to be rather

weak on its own, since this sole interfacial mutation in

Omicron B.1.1.529 cannot prevent P2C5 from efficient-

ly recognizing the corresponding RBD (Table1). In this

respect, P2C5 surpasses a recently described neutral-

izing nanobody 2S-1-19 in that the latter had lost the

ability to recognize Omicron BA.1 (the same mutations

as in B.1.1.529) [19]. Mutation of this Glu484 to lysine

in Gamma RBD apparently is dispensable for the P2C5

binding as well, especially since the charge reversal

at the 484 position in Gamma RBD (E484K) can make

an additional, weak salt bridge with Glu46 of P2C5

(Fig.4b). By exclusion, this indicates that the key role

in destabilizing the RBD:P2C5 interaction in the case

of Omicron XBB.1 RBD is played by the F490S muta-

tion (Fig. 4,a-c). Indeed, as mentioned, Phe490 is lo-

cated right in the hydrophobic cluster stabilizing the

RBD:P2C5 interface and its side chain is stacked with

the opposite Phe47 side chain of P2C5, whereas the

removal of the aromatic Phe490 functionality would

clearly compromise the cluster stability, especially ac-

companied by the introduction of the short polar side

chain of a serine (Fig. 4c). Most delightfully, the role

of other, numerous mutations found in the Omicron

XBB.1 variant RBD can be neglected since all those

are located beyond the region directly involved in the

formation of the RBD:P2C5 interface, which fortunate-

ly excludes the necessity of conducting sophisticated

combinatorial biochemical assays for assignment of

the contribution of these mutations.

To sum up, our structural data explain the likely

mechanism of the neutralization activity of the P2C5

nanobody on most of the earlier SARS-CoV-2 variants

circulating before the emergence of Delta. Further-

more, the crystal structure obtained has allowed us to

narrow the extended list of more recently emerged mu-

tations, for example, found in the Delta and Omicron

XBB.1 variants, potentially affecting the neutralization

activity of P2C5, down to just two mutations, L452R

and F490S, which rationalize the evasion of the Delta

and Omicron XBB.1 variants of RBD from the action

of P2C5. Interestingly, these mutations interfere with

the RBD recognition by other nanobodies including the

recently reported 2S-1-19 nanobody [19], despite the

appreciably different footprints on RBD of this nano-

body and P2C5 (Fig.5). Last but not least, we expect

that the crystal structure obtained can support P2C5

reengineering aimed at compensation for the effects

of the new RBD mutations [7] and also can facilitate

mapping of epitopes of other neutralizing antibod-

ies against RBD in competition assays, even without

requiring determination of new structures.

Contributions. N.N.S., D.V.S., and D.Y.L. initiat-

ed the project, I.A.F., I.V.D., A.I.K., I.A.A., I.B.E., A.A.D.,

V.V.P., and I.D.Z. obtained RBD variants and antibody,

performed functional, in vitro and neutralization

tests, N.N.S. prepared and purified complexes for

SLUCHANKO et al.1270

BIOCHEMISTRY (Moscow) Vol. 89 No. 7 2024

Fig. 5. Comparison of the RBD binding modes for nanobodies P2C5 (this work) and 2S-1-19 (PDB ID 8H91 [19]). The two crystal

structures were superimposed by aligning RBD so that the different orientation and binding footprints are seen. The two projec-

tions are presented. Location of the residues Leu452 and Phe490 mutated in Delta and Omicron XBB variants are indicated. The

residues Leu455-Phe456, mutated in newest SARS-CoV-2 variants to Phe455-Leu456 and thereby knocking out many antibodies

[6], are shown by blue spheres representing their Cα atoms. Of note, P2C5 antibody binds from the other side of RBD and should

not be sensitive to these mutations.

crystallization, L.A.V. crystallized protein complex,

L.A.V. and K.M.B. collected X-ray data, N.N.S. solved

structures, N.N.S. and L.A.V. refined structures, N.N.S.,

L.A.V., andK.M.B. validated structures, N.N.S. analyzed

and compared structures, wrote the paper with input

from all authors, D.V.S., D.Y.L., A.L.G., and V.O.P super-

vised the study, V.O.P. acquired funding.

Funding. The study was financially supported

by the Russian Science Foundation (project no.23-74-

30004).

Ethics declarations. This work does not con-

tain any studies involving human and animal sub-

jects. Theauthors of this work declare that they have

noconflicts of interest.

Open access. This article is licensed under a Cre-

ative Commons Attribution 4.0 International License,

which permits use, sharing, adaptation, distribution,

and reproduction in any medium or format, as long

as you give appropriate credit to the original author(s)

and the source, provide a link to the Creative Commons

license, and indicate if changes were made. Theimages

or other third-party material in this article are includ-

ed in the article’s Creative Commons license, unless

indicated otherwise in a credit line to the material.

Ifmaterial is not included in the article’s Creative Com-

mons license and your intended use is not permitted

by statutory regulation or exceeds the permitted use,

you will need to obtain permission directly from the

http://creativecommons.org/licenses/by/4.0/.

REFERENCES

1. Kansagra,K., Parmar, D., Mendiratta, S.K., Patel,J.,

Joshi,S., Sharma,N., Parihar,A., Bhoge,S., Patel,H.,

Kalita,P., Munshi,R., Kurmi, P., Shah, R., Gupta,A.,

Bhalla,H., Bekkalele,H., Verma,R., Agarwal,D., Shar-

ma,S., Gawande,A., etal. (2021) A phase 3, random-

ized, open-label, noninferiority trial evaluating an-

ti-rabies monoclonal antibody cocktail (TwinrabTM)

Against Human Rabies Immunoglobulin (HRIG), Clin.

Infect. Dis., 73, e2722-e2728, https://doi.org/10.1093/cid/

ciaa779.

2. Hammitt, L. L., Dagan, R., Yuan, Y., Baca Cots, M.,

Bosheva, M., Madhi, Sh. A., Muller, W. J., Zar, H. J.,

Brooks,D., Grenham,A., Wählby, H.U., Mankad, V.S.,

Ren, P., Takas, T., Abram, M. E., Leach, A., Griffin,

STRUCTURE OF THE SARS-CoV-2 RBD:P2C5 NANOBODY COMPLEX 1271

BIOCHEMISTRY (Moscow) Vol. 89 No. 7 2024

M.P., and Villafana,T. (2022) Nirsevimab for preven-

tion of RSV in healthy late-preterm and term infants,

New Eng. J.Med., 386, 837-846, https://doi.org/10.1056/

NEJMoa2110275.

3. Mulangu,S., Dodd Lori,E., Davey Richard,T., Tshiani

Mbaya, O., Proschan, M., Mukadi, D., Lusakibanza,

M.M., Nzolo, D., Tshomba, O.A., Ibanda,A., Ali, R.,

Coulibaly,S., Levine Adam,C., Grais,R., Diaz,J., Lane,

H.C., and Muyembe-Tamfum, J.-J. (2019) A random-

ized, controlled trial of Ebola virus disease thera-

peutics, New Eng. J. Med., 381, 2293-2303, https://

doi.org/10.1056/NEJMoa1910993.

4. Wang,Q., and Zhang,L. (2020) Broadly neutralizing

antibodies and vaccine design against HIV-1 infection,

Front. Med., 14, 30-42, https://doi.org/10.1007/s11684-

019-0721-9.

5. Jarmo,O., Veli-Jukka,A., and Eero, M. (2020) Treat-

ment of Clostridioides (Clostridium) difficile infection,

Ann. Med., 52, 12-20, https://doi.org/10.1080/07853890.

2019.1701703.

6. Liu,C., Das,R., Dijokaite-Guraliuc,A., Zhou,D., Men-

tzer, A.J., Supasa,P., Selvaraj,M., Duyvesteyn, H.M.E.,

Ritter, T.G., Temperton,N., Klenerman,P., Dunachie,

S.J., Paterson, N.G., Williams, M.A., Hall, D.R., Fry,

E.E., Mongkolsapaya,J., Ren,J., Stuart, D.I., and Scre-

aton, G. R. (2024) Emerging variants develop total

escape from potent monoclonal antibodies induced

by BA.4/5 infection, Nat. Commun., 15, 3284, https://

doi.org/10.1038/s41467-024-47393-3.

7. Laroche,A., Orsini Delgado, M.L., Chalopin,B., Cuni-

asse,P., Dubois,S., Sierocki,R., Gallais,F., Debroas,S.,

Bellanger,L., Simon,S., Maillère,B., and Nozach,H.

(2022) Deep mutational engineering of broadly-neu-

tralizing nanobodies accommodating SARS-CoV-1 and

2 antigenic drift, mAbs, 14, 2076775, https://doi.org/

10.1080/19420862.2022.2076775.

8. Hannula,L., Kuivanen,S., Lasham,J., Kant,R., Karein-

en,L., Bogacheva,M., Strandin,T., Sironen,T., Hepojo-

ki,J., Sharma,V., Saviranta,P., Kipar,A., Vapalahti,O.,

Huiskonen, J. T., and Rissanen, I. (2024) Nanobody

engineering for SARS-CoV-2 neutralization and detec-

tion, Microbiol. Spectrum, 12, e04199-04122, https://

doi.org/10.1128/spectrum.04199-22.

9. Hanke, L., Vidakovics Perez, L., Sheward, D. J.,

Das,H., Schulte,T., Moliner-Morro,A., Corcoran,M.,

Achour,A., Karlsson Hedestam, G.B., Hällberg, B.M.,

Murrell, B., and McInerney, G. M. (2020) An alpaca

nanobody neutralizes SARS-CoV-2 by blocking re-

ceptor interaction, Nat. Commun., 11, 4420, https://

doi.org/10.1038/s41467-020-18174-5.

10. Hoffmann, M., Kleine-Weber, H., Schroeder, S.,

Krüger,N., Herrler,T., Erichsen,S., Schiergens, T.S.,

Herrler, G., Wu, N.-H., Nitsche, A., Müller, M. A.,

Drosten, C., and Pöhlmann, S. (2020) SARS-CoV-2

cell entry depends on ACE2 and TMPRSS2 and is

blocked by a clinically proven protease inhibitor,

Cell, 181, 271-280.e278, https://doi.org/10.1016/j.cell.

2020.02.052.

11. Lan, J., Ge, J., Yu, J., Shan, S., Zhou, H., Fan, S.,

Zhang,Q., Shi,X., Wang,Q., Zhang,L., and Wang,X.

(2020) Structure of the SARS-CoV-2 spike recep-

tor-binding domain bound to the ACE2 receptor,

Nature, 581, 215-220, https://doi.org/10.1038/s41586-

020-2180-5.

12. Piccoli,L., Park, Y.-J., Tortorici, M.A., Czudnochows-

ki,N., Walls, A.C., Beltramello,M., Silacci-Fregni,C.,

Pinto, D., Rosen, L. E., Bowen, J. E., Acton, O. J.,

Jaconi,S., Guarino,B., Minola,A., Zatta,F., Sprugas-

ci,N., Bassi,J., Peter,A., De Marco,A., Nix, J.C., etal.

(2020) Mapping neutralizing and immunodominant

sites on the SARS-CoV-2 spike receptor-binding do-

main by structure-guided high-resolution serology,

Cell, 183, 1024-1042.e1021, https://doi.org/10.1016/

j.cell.2020.09.037.

13. Xiang, Y., Nambulli, S., Xiao, Z., Liu, H., Sang, Z.,

Duprex, W. P., Schneidman-Duhovny, D., Zhang, C.,

and Shi, Y. (2020) Versatile and multivalent nano-

bodies efficiently neutralize SARS-CoV-2, Science,

370, 1479-1484, https://doi.org/10.1126/science.

abe4747.

14. Esmagambetov, I. B., Ryabova, E.I., Derkaev, A.A.,

Shcheblyakov, D. V., Dolzhikova, I. V., Favorskaya,

I.A., Grousova, D.M., Dovgiy, M.A., Prokofiev, V.V.,

Gosudarev, A.I., Byrikhina, D.V., Zorkov, I.D., Iliukhi-

na, A.A., Kovyrshina, A.V., Shelkov, A.Y., Naroditsky,

B.S., Logunov, D.Y., and Gintsburg, A.L. (2023) rAAV

expressing recombinant antibody for emergency

prevention and long-term prophylaxis of COVID-19,

Front. Immunol., 14, 1129245, https://doi.org/10.3389/

fimmu.2023.1129245.

15. Favorskaya, I.A., Shcheblyakov, D. V., Esmagambe-

tov, I. B., Dolzhikova, I. V., Alekseeva, I. A., Korob-

kova, A.I., Voronina, D.V., Ryabova, E. I., Derkaev,

A. A., Kovyrshina, A. V., Iliukhina, A. A., Botikov,

A.G., Voronina, O.L., Egorova, D.A., Zubkova, O.V.,

Ryzhova, N.N., Aksenova, E.I., Kunda, M.S., Logun-

ov, D.Y., Naroditsky, B.S., etal. (2022) Single-domain

antibodies efficiently neutralize SARS-CoV-2 vari-

ants of concern, Front. Immunol., 13, 822159, https://

doi.org/10.3389/fimmu.2022.822159.

16. Deshpande, A., Harris, B. D., Martinez-Sobrido, L.,

Kobie, J.J., and Walter, M.R. (2021) Epitope classifi-

cation and RBD binding properties of neutralizing

antibodies against SARS-CoV-2 variants of concern,

Front. Immunol., 12, 691715, https://doi.org/10.3389/

fimmu.2021.691715.

17. Jian,F., Feng,L., Yang,S., Yu,Y., Wang,L., Song,W.,

Yisimayi,A., Chen,X., Xu,Y., Wang,P., Yu,L., Wang,J.,

Liu, L., Niu, X., Wang, J., Xiao, T., An, R., Wang, Y.,

Gu,Q., Shao,F., etal. (2023) Convergent evolution of

SARS-CoV-2 XBB lineages on receptor-binding domain

455-456 synergistically enhances antibody evasion

SLUCHANKO et al.1272

BIOCHEMISTRY (Moscow) Vol. 89 No. 7 2024

and ACE2 binding, PLOS Pathogens, 19, e1011868,

https://doi.org/10.1371/journal.ppat.1011868.

18. Yang, J., Lin, S., Sun, H., Chen, Z., Yang, F., Lin, X.,

Guo,L., Wang,L., Wen,A., Zhang,X., Dai,Y., He,B.,

Cao,Y., Dong,H., Liu,X., Chen,B., Li,J., Zhao,Q., and

Lu, G. (2022) A potent neutralizing nanobody tar-

geting the spike receptor-binding domain of SARS-

CoV-2 and the structural basis of its intimate binding,

Front. Immunol., 13, 820336, https://doi.org/10.3389/

fimmu.2022.820336.

19. Zhang, Y., Wang, D., Xiang, Q., Hu, X., Zhang, Y.,

Wu,L., Zhang,Z., Wang,Y., Zhao,J., McCormick, P.J.,

Fu,J., Fu,Y., Zhang,J., Jiang,H., and Li, J. (2024) A

potent neutralizing nanobody targeting a unique epi-

tope on the receptor-binding domain of SARS-CoV-2

spike protein, Virology, 589, 109925, https://doi.org/

10.1016/j.virol.2023.109925.

20. Boyko, K. M., Varfolomeeva, L. A., Egorkin, N. A.,

Minyaev, M. E., Alekseeva, I. A., Favorskaya, I. A.,

Ryabova, E.I., Prokofiev, V.V., Esmagambetov, I.B.,

Shcheblyakov, D.B., Logunov, D.Y., Gintsburg, A.L.,

Popov, V. O., and Sluchanko, N. N. (2023) Prepa-

ration and crystallographic analysis of a complex

of SARS-CoV-2 S-protein receptor-binding domain

with a virus-neutralizing nanoantibody, Crys-

tallogr. Rep., 68, 864-871, https://doi.org/10.1134/

S1063774523601168.

21. Vagin, A., and Teplyakov, A. (2010) Molecular re-

placement with MOLREP, Acta Crystallogr. D Biol.

Crystallogr., 66, 22-25, https://doi.org/10.1107/

S0907444909042589.

22. Jumper,J., Evans,R., Pritzel,A., Green,T., Figurnov,M.,

Ronneberger, O., Tunyasuvunakool, K., Bates, R.,

Zidek, A., Potapenko, A., Bridgland, A., Meyer, C.,

Kohl, S.A.A., Ballard, A.J., Cowie,A., Romera-Pare-

des, B., Nikolov, S., Jain, R., Adler, J., Back, T., et al.

(2021) Highly accurate protein structure predic-

tion with AlphaFold, Nature, 596, 583-589, https://

doi.org/10.1038/s41586-021-03819-2.

23. Murshudov, G. N., Skubák, P., Lebedev, A. A., Pan-

nu, N.S., Steiner, R. A., Nicholls, R.A., Winn, M.D.,

Long, F., and Vagin, A. A. (2011) REFMAC5 for the

refinement of macromolecular crystal structures,

Acta Cryst. D, 67, 355-367, https://doi.org/10.1107/

S0907444911001314.

24. Emsley,P., and Cowtan,K. (2004) Coot: model-build-

ing tools for molecular graphics, Acta Crystallogr. D

Biol. Crystallogr., 60, 2126-2132, https://doi.org/10.1107/

S0907444904019158.

25. Shi,Z., Li,X., Wang,L., Sun,Z., Zhang,H., Chen,X.,

Cui,Q., Qiao,H., Lan,Z., Zhang,X., Li,X., Li,L., Xu,J.,

Gong,R., Fan,C., and Geng, Y. (2022) Structural ba-

sis of nanobodies neutralizing SARS-CoV-2 variants,

Structure, 30, 707-720.e705, https://doi.org/10.1016/

j.str.2022.02.011.

26. Park, Y.-J., Pardon,E., Wu, M., Steyaert, J., and Hol,

W. G. J. (2012) Crystal structure of a heterodimer

of editosome interaction proteins in complex with

two copies of a cross-reacting nanobody, Nucleic

Acids Res., 40, 1828-1840, https://doi.org/10.1093/

nar/gkr867.

27. Abramson,J., Adler,J., Dunger,J., Evans,R., Green,T.,

Pritzel, A., Ronneberger, O., Willmore, L., Ballard,

A. J., Bambrick, J., Bodenstein, S. W., Evans, D. A.,

Hung, C.-C., O’Neill, M., Reiman, D., Tunyasuvuna-

kool, K., Wu, Z., Žemgulytė, A., Arvaniti, E., Beat-

tie, C., et al. (2024) Accurate structure prediction

of biomolecular interactions with AlphaFold 3, Na-

ture, 630, 493-500, https://doi.org/10.1038/s41586-

024-07487-w.

28. Krissinel, E., and Henrick, K. (2007) Inference of

macromolecular assemblies from crystalline state,

J. Mol. Biol., 372, 774-797, https://doi.org/10.1016/

j.jmb.2007.05.022.

Publisher’s Note. Pleiades Publishing remains

neutral with regard to jurisdictional claims in pub-

lished maps and institutional affiliations.