Artículo Micro Médica

Subscriber access provided by Universiteit Utrecht
Article
Searching for antipneumococcal targets: Choline-
binding modules as phagocytosis enhancers
Emma Roig-Molina, Manuel Sánchez-Angulo, Jana Seele, Francisco
García-Asencio, Roland Nau, Jesús Sanz, and Beatriz Maestro
ACS Infect. Dis., Just Accepted Manuscript • DOI: 10.1021/acsinfecdis.9b00344 • Publication Date (Web): 05 Mar 2020
Downloaded from pubs.acs.org on March 9, 2020
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,

Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the
course of their duties.
Page 1 of 52 ACS Infectious Diseases
1
2
3
Searching for antipneumococcal targets:
4
5 Choline-binding modules as phagocytosis enhancers
6
7
8
9
10 Emma Roig-Molina†, Manuel Sánchez-Angulo‡, Jana Seele§,∥,
11
12
13
Francisco García-Asencio†,⊥, Roland Nau§,∥, Jesús M. Sanz†,#,‡,∇,* and
14
15 Beatriz Maestro†,#,∇,*
16
17
18
19 †
20
Institute of Research, Development, and Innovation in Healthcare Biotechnology in
21 Elche (IDiBE). Miguel Hernández University. Avda Universidad s/n. Elche-03202
22 (Spain)
23
24
‡Department of Vegetal Production and Microbiology. Miguel Hernández University.
25
26 Avda Universidad s/n. Elche-03202 (Spain)
27
28
§
29 Department of Geriatrics, Evangelisches Krankenhaus Göttingen-Weende, An der
30 Lutter 24, 37075 Göttingen (Germany)
31
32
∥Department of Neuropathology, University Medical Center Göttingen, Georg-August-
33
34 University Göttingen, Wilhelmsplatz 1, 37073 Göttingen (Germany)
35
36 #
37 Biological Research Centre, Spanish National Research Council (CSIC), c/ Ramiro de
38 Maeztu, 9, Madrid-28040, Spain.
39
40 ‡
41 Centro de Investigación Biomédica en Red de Enfermedades Respiratorias (CIBERES),
42 Madrid-28040, Spain.
43
44
⊥Present address: Developmental Neurobiology Unit, Institute of Neurosciences, Miguel
45
46 Hernández University - Spanish National Research Council (UMH-CSIC), Av. Ramón y
47 Cajal s/n, Sant Joan d'Alacant-03550, Spain
48
49
50 * Corresponding authors: jmsanz@cib.csic.es, bmaestro35@gmail.com
51
52
∇ J.M.S. and B.M. contributed equally to this work.
53
54
55
56
57
58
59
60
1
ACS Paragon Plus Environment
ACS Infectious Diseases Page 2 of 52
1
2
3 Choline-binding proteins (CBPs) from Streptococcus pneumoniae comprise a family of
4
5 modular polypeptides involved in essential events of this pathogen. They recognize the
6
7
choline residues present in the teichoic and lipoteichoic acids of the cell wall using the
8
so-called choline-binding modules (CBMs). The importance of CBPs in pneumococcal
9
10 physiology points to them as novel targets to combat antimicrobial resistances shown by
11
12 this organism. In this work we have tested the ability of exogenously added CBMs to act
13
14 as CBP inhibitors by competing with the latter for the binding to the choline molecules
15
16 in the bacterial surface. First, we carried out a thorough physicochemical characterization
17
18 of three native CBMs, namely C-LytA, C-Cpl1 and C-CbpD, and assessed their affinity
19
20 for choline and macromolecular, pneumococcal cell-wall mimics. The interaction with
21
22 these substrates was evaluated by molecular modeling, analytical ultracentrifugation,
23
24 surface plasmon resonance and fluorescence and circular dichroism spectroscopies. Van't
25
26 Hoff thermal analyses unveiled the existence of one non-canonical choline binding site
27
28 in each of the C-Cpl1 and C-CbpD proteins, leading in total to 5 ligand-binding sites per
29
30
dimer and 4 sites per monomer, respectively. Remarkably, the binding affinities of the
31
CBMs do not directly correlate with their native oligomeric state or with the number of
32
33 choline-binding sites, suggesting that choline recognition by these modules is a complex
34
35 phenomenon. On the other hand, the exogenous addition of CBMs to pneumococcal
36
37 planktonic cultures caused extensive cell-chaining probably as a consequence of the
38
39 inhibition of CBP attachment to the cell wall. This was accompanied by bacterial
40
41 aggregation and sedimentation, causing an enhancement of bacterial phagocytosis by
42
43 peritoneal macrophages. In addition, the rational design of an oligomeric variant of a
44
45 native CBM led to a substantial increase in its antibacterial activity by multivalency
46
47 effects. These results suggest that CBMs might constitute promising non-lytic
48
49 antimicrobial candidates based on the natural induction of the host defense system.
50
51
52 KEYWORDS Streptococcus pneumoniae, choline-binding proteins, magnetic
53
54
55 nanoparticles, antimicrobial resistance, DEAE, bacterial cell wall
56
57
58
59
60
2
1
2 INTRODUCTION
3
4 Streptococcus pneumoniae (pneumococcus) is a Gram-positive bacterium
5
6
7 responsible for acute life-threatening infections including meningitis and sepsis,1 and it
8
9 constitutes the most frequently detected bacterial pathogen in cases of community-
10
11 acquired pneumonia.2 Moreover, it is a major causative agent of otitis media and
12
13
14 sinusitis.3 Pneumococcal diseases are widespread both in industrialized and in developing
15
16 countries, leading to about 1.2 million deaths per year according to the World Health
17
18 Organization (WHO), 70% of which correspond to children under 5 and elderly adults
19
20
21
above 70 years old.4 Furthermore, S. pneumoniae is considered by the WHO as a priority
22
23 pathogen for R&D of new antibiotics5 and as "serious" threat by the U.S. Department of
24
25 Health.6
26
27
28 Current intervention strategies against pneumococcal disease, based on
29
30
31
vaccination and antibiotic treatment, face several drawbacks. Present-day vaccination
32
33 only covers part of the nearly 100 serotypes already described for this pathogen,7 and the
34
35 immunogenic effect of the available vaccines may be limited in some cases,8 while
36
37
serotype replacement phenomena are also frequently reported9-10 leading to resurgent
38
39
40 disease.11 Besides, vaccines do not always protect in developing countries, either because
41
42 non-vaccine serotypes predominate, or due to insufficient access to the vaccination
43
44 programmes.12-13 On the other hand, antimicrobial resistance constitutes a serious
45
46
47 concern: appearance of new non-susceptible pneumococcal strains to β-lactam and/or
48
49 macrolid antibiotics are commonly reported,14-15 reaching resistance levels of c.a. 50 %
50
51 in some European countries.16 This situation makes necessary the research on new
52
53
54
effective chemotherapies involving distinctive mechanisms.
55
56
57 The cell wall of S. pneumoniae, and especially the proteins attached to its surface,
58
59 constitute an attractive antimicrobial target to develop new drugs. Three main groups of
60
3
1
2 surface proteins in pneumococcus have been described, namely i) lipoproteins, ii)
3
4 choline-binding proteins (CBPs) and iii) LPXTG-sequence-containing proteins, the latter
5
6
7 being covalently attached to the peptidoglycan.17 Among them, CBPs represent a family
8
9 of 13-16 members, depending on the strain, that are involved in several essential
10
11 processes of this pathogen, e.g. cell-wall hydrolysis prior to daughter cell separation upon
12
13
14 division and release of virulence factors (LytA, LytB, LytC), adhesion to the host (PspC),
15
16 bacterial fratricide (LytA, LytC, CbpD) or evasion from the immune system (LytA, PspA,
17
18 PspC).18 Furthermore, these proteins are highly conserved among all serotypes and
19
20
21
recognize the choline moieties present in the cell wall teichoic and lipoteichoic acids,
22
23 allowing their non-covalent adsorption to the macromolecular peptidoglycan structure by
24
25 means of a choline-binding module (CBM).18 The CBMs differ, among themselves, in
26
27 size, oligomeric state and position in the protein (N- or C-termini), but all of them consist
28
29
30 of a variable number of aromatic-rich motifs, about 20 residues long, termed choline-
31
32 binding repeats (CBRs). These CBRs belong to the widespread CW_binding 1_motif
33
34 family (PFAM code PF01473) (http://pfam.xfam.org/family/PF01473) and are
35
36
37 structurally formed by a ≈ 14-aa β-hairpin followed by a ≈ 6-aa linker. The choline-
38
39 binding modules are three-dimensionally arranged in a β,β-3 solenoid superstructure,
40
41 generating a choline-binding site between any two consecutive CBRs and in which the
42
43
44
ligand is primarily stabilized by cation-π and van der Waals interactions with the aromatic
45
46 side chains.18
47
48
49 Taking into account the variety of roles carried out by the CBPs (as involved in
50
51 all key aspects of the pneumococcal vital cycle), several efforts have been carried out to
52
53
54
inhibit choline recognition and hence to displace CBPs from the cell surface.
55
56 Exogenously-added free choline is a long known CBP inhibitor, competing with those
57
58 molecules present in the teichoic acids on binding to the protein.19-20 As a result, daughter
59
60
cell separation is hampered, leading to the appearance of long cellular chains.20 However,
4
1
2 the poor affinity of CBPs for the free ligand21-25 makes necessary the use of millimolar
3
4 concentrations of this compound, which are therapeutically not achievable. Nevertheless,
5
6
7 choline binding affinity to CBPs increases exponentially by using multivalent
8
9 nanoparticles such as choline dendrimers which inhibit the CBPs activity in vitro around
10
11 10,000-fold more efficiently than free choline, interfering with daughter cell separation
12
13
14 after division26 and favouring phagocytosis of the cell chains by microglial cells.27
15
16
17 Induction of host phagocytosis may constitute an interesting way to find novel
18
19 therapies. Some widely used molecules, for instance lytic β-lactam antibiotics, have been
20
21 shown to induce the release of inflammatory and damaging toxins such as pneumolysin
22
23
24 in contrast to non-lytic antimicrobials. The use of non-lytic antibacterials appears to be
25
26 beneficial in experimental models in comparison with β-lactam antibiotics. Limited
27
28 clinical data suggest that this approach may be also beneficial in human infections with a
29
30
31
high bacterial load.28-29 Moreover, release of intracellular content by autolysis has been
32
33 described to interfere with the normal host phagocytic system.30 This prompted us to
34
35 further investigate alternative mechanisms of CBP inhibition that do not involve cell lysis.
36
37
In this sense, we had previously found that complexing the choline residues in the cell
38
39
40 wall with isolated CBMs such as C-LytA and C-Cpl1 (derived from the bacterial autolysin
41
42 LytA and the phagic lysozyme Cpl1, respectively) resulted in the inhibition of the
43
44 enzymatic activity in vitro of their full parental enzymes (LytA and Cpl1) which then
45
46
47 would not be capable to reach their targets.31
48
49
50 Based on these preliminary results, we decided to further develop the idea of using
51
52 isolated CBMs as non-lytic antipneumococcal molecules to bind to the choline residues
53
54 on the pneumococcal surface, making them unable to be recognized by parental, full-
55
56
57 length CBPs. Our strategy is shown in Figure 1. Among other effects, CBM-mediated
58
59 inhibition of peptidoglycan hydrolases such as LytA and LytB should hamper daughter
60
5
1
2 cell separation upon division, leading to chain formation and eventually favouring the
3
4 phagocytosis of the chains, similarly to addition of choline and choline dendrimers.27, 32
5
6
7 To this aim we have carried out a biophysical characterization of three CBMs differing
8
9 in number of CBRs and oligomeric state, namely C-LytA, C-CbpD and C-Cpl1, which
10
11 belong to the bacterial host LytA and CbpD proteins and the pneumococcal Cpl-1 phage-
12
13
14 encoded lysozyme, respectively. The LytA amidase is the major murein hydrolase of S.
15
16 pneumoniae and its C-terminal module (C-LytA) the best studied species of the CBM
17
18 family22, 33-34 (PDB code: 1HCX). On the other hand, CbpD is a murein hydrolase that
19
20
21
assists in cellular fratricide together with LytA, LytC and CibAB.35. Finally, Cpl1 is the
22
23 lysozyme synthesized by Cp-1 phage, and it is responsible of bacterial lysis to release the
24
25 phage progeny.36 Only the three-dimensional structure of the full parental enzyme is
26
27 known37-38(PDB codes: 1H09 and 1OBA).
28
29
30 Functional module
Ch: Choline residue
31 Choline-binding module
32
CBP
33
34
35
36 Cell CBP Cell wall
37 division binding hydrolisis
38 Diplococcal shape
39
40
41
42
43 Cell
Further ...
44
division
divisions
45 Chaining → Phagocytosis 
46
47 Figure 1. Strategy for the use of CBMs as non-lytic antipneumococcal molecules. Under usual growth
48 conditions, cells divide and then activate full-length CBP hydrolases such as LytA and LytB that bind to
49 choline residues in the teichoic acids and carry out the selective hydrolysis of the cell wall leading to the
50 separation of daughter cells. On exogenous addition of CBMs, CBP hydrolases cannot attach to the cell wall
51
as all choline sites are occupied, so daughter cells do not separate upon division, leading to cell chaining
52
53 which hampers infectivity and promoting pathogen removal by phagocytosis.
54
55
56
57 In this work we have assessed the affinity of these CBMs for choline and synthetic
58
59 cell-wall mimics with a variety of biophysical methods. Furthermore, we studied the
60
6
1
2 effects, as possible CBP inhibitors, of externally added CBMs on planktonic cultures of
3
4 S. pneumoniae and finally on the induction of phagocytosis of bacterial cells by mouse
5
6
7 peritoneal macrophages. Our results confirm that inhibition of adsorption of CBPs to the
8
9 cell wall may pave the way for the development of novel, non-lytic antipneumococcal
10
11 therapies.
12
13
14
15
16
17
18
RESULTS
19
20 Description of choline-binding modules. The C-LytA polypeptide (as coded by
21
22 the expression plasmid pCE17)31 contains 5 CBRs (corresponding to CBRs 2-6 of the
23
24
full-length LytA amidase) plus a hairpin tail (Figure 2A) configuring 4 choline-binding
25
26
27 sites of either low and high affinity.23. On choline binding, dimerization occurs through
28
29 the C-terminal hairpin, adopting a “boomerang-like” shape.33, 39 With regard to C-Cpl1,
30
31 it is made up of 6 CBRs highly similar in sequence to those of LytA40(Figure 2A) although
32
33
34 there are important structural differences among them: C-Cpl1 only contains 2 choline-
35
36 binding sites, compared to of 4 ligands for C-LytA.33 Moreover, although both proteins
37
38 acquire a similar “boomerang-like” dimeric shape upon binding of choline, C-Cpl1
39
40
41
dimerizes through the N-terminal part, and probably configures an additional choline-
42
43 binding site at the dimer interface.41
44
45
46 The three-dimensional structure of C-CbpD is unknown, although sequence
47
48 comparison analyses predicts the presence of four CBRs, therefore constituting the
49
50
51 smallest representative of pneumococcal CBMs described so far18, 42 Taking advantage of
52
53 the abundance of similar CBMs with an elucidated tertiary structure, we modeled the
54
55 structure of C-CbpD with the SwissModel homology procedure (Figure 3).43 Among all
56
57
identified targets, the optimal template was the C-terminal part of the choline-binding
58
59
60 protein CbpJ.43 According to the model, the protein acquires the typical β-solenoid
7
1
2 disposition of CBRs which is characteristic of the CBP family, with the additional
3
4 contribution of a C-terminal β-hairpin comprising residues 434-441. Three canonical
5
6
7 choline-binding sites are observed in the model, each containing three aromatic residues
8
9 able to establish direct contacts with the ligand, plus a fourth, inner aromatic side chain
10
11 stabilizing the cavity through T-shaped π-π interactions with the central aromatic residue
12
13
14 (Figure 3A)18. Furthermore, a fourth non-standard binding site involving Tyr 439 in the
15
16
17
18 A 188
·
200
·
220
·
MKGGIVHSDGSYPKDKFEKINGTWYYFDSSGYMLADRWRK
19 240 260
20 C-LytA ·
HTDGNWYWFDNSGEMATGWKKIADKWYYFNEEGAMKTGWV
·
21 2 3 4 5 6 β 280
·
KYKDTWYYLDAKEGAMVSNAFIQSADGTGWYYLKPDGTLA
300
·
22 CBRs
318
23
·
DRPEFTVEPDGLITVK
24 185
·
200
·
25 MKGGIPIVLLDDEEDDKPKTAGTWKQDSKGWWFRRNNGSF
26 2
C-Cpl1
220
·
240
·
PYNWEKIGGVWYYFDSKGYCLTSEWLKDNEKWYYLKDNGA
27 1 2 3 4 5 6 β 260 280
·
28
·
MATGWVLVGSEWYYMDDSGAMVTGWVKYKNNWYYMTNERG
CBRs
29 300
·
320
·
338
·
NMVSNEFIKSGKGWYFMNTNGELADNPSFTKEPDGLITVA
30
31
353 360 380
32 · ·
MGWKKINGSWYHFKSNGSKSTGWLKDGSSWYYLKLSGEMQ
·
33 C-CbpD
2 400
·
420
·
34 1 2 3 4 β TGWLKENGSWYYLGSSGAMKTGWYQVSGEWYYSYSSGALA
35
440 448
CBRs · ·
INTTVDGYRVNSDGERV
36
37
38
B
39
40
41
42
43
44
45
46
47 C-CbpD C-LytA C-Cpl1 C-CbpD
C-LytA C-Cpl1
48
49
Figure 2. A) Amino acid sequence of CBMs used in this study, indicating the numbered CBRs and the
50
51 C-terminal tail, together with a secondary structure representation. Choline or choline analogs in the
52 binding sites are displayed in stick representation. Residues in italics do not belong to the native parental
53 sequence. CBR and amino acid numbering scheme is that of the parental CBP. B) Three-dimensional
54 surface representation with coulombic colouring (red indicates negative polarity, blue indicates positive
55 polarity, white indicates zero polarity). C-LytA and C-Cpl1 are represented by their crystallographic
56 structures (PDB codes 1HCX and 1OBA, respectively) whereas C-CbpD is represented by a homology
57 model with docked choline molecules (see text). Figures were rendered with UCSF Chimera 1.10 74.
58
59
60
8
1
2 C-terminal hairpin might also be configured, as described for CbpJ44. To assess whether
3
4 the geometry of these cavities might be suitable for choline binding, we studied their
5
6
7 interaction in silico with choline molecules by docking procedures using the SwissDock
8
9 server.45 As depicted in Figure 3A, the four predicted sites would nicely accommodate
10
11 the ligands through cation-π and van der Waals interactions. Remarkably, the non-
12
13
14 canonical site located at the C-terminus would be further stabilized by an additional π-π
15
16 contact of Tyr 423 with Tyr 425, also present in the C-terminal tail, leading to a notable
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51 Figure 3. Three-dimensional structural model of C-CbpD generated by the SwissModel and SwissDock
52 servers (see Experimental Section for details). A, overall secondary structure representation, showing in
53 stick format the aromatic residues configuring the predicted choline-binding sites (those involved in
54 secondary T-shaped π-π interactions are labeled in italics). Choline ligands are also shown in stick
55 representation, and were generated by docking methods; B, a closer view of the predicted, non-standard
56 choline binding site configured by the C-terminal β-hairpin and the following tag. Aromatic residues
57 involved either in direct interaction with the ligand or in the stabillization of the site are shown as van
58 der Waals spheres.
59
60
9
1
2 network of aromatic residues disposed in a T-shaped stabilizing conformation (Figure
3
4 3B).
5
6
7
8
Oligomeric state of CBMs. To carry out the thermodynamic characterization of
9
10 the binding of choline to CBMs, the oligomerization state of each protein at any ligand
11
12 concentration should be unequivocally defined. In case of C-LytA, binding affinity
13
14
constants have been thoroughly determined22 (Table 1) and it has been demonstrated that
15
16
17 the protein is a dimer above 10 mM choline.23 The C-Cpl1 protein becomes a dimer at 40
18
19 mM choline21, 41
but there is a lack of information at lower concentrations, and the
20
21 oligomeric state of C-CbpD has not been studied at all so far. Therefore, to check the
22
23
24 oligomeric state of C-Cpl1 and C-CbpD in the absence or presence of 10 mM choline (a
25
26 concentration often used in our experiments, see below), sedimentation velocity
27
28 experiments were carried out. Results for C-Cpl1 (Figure 4) in the absence of choline
29
30
31
show a single species (>99 %) with a sedimentation coefficient of 1.87 S and a frictional
32
33 ratio (f/fo) of 1.33, corresponding to the monomer and in agreement with the previously
34
35 reported results for this protein (1.81 S).41 In the presence of the ligand, a 2.52 S major
36
37
species was detected accounting for around 86 % of total signal and a frictional ratio (f/fo)
38
39
40 of 1.6, which is compatible with a slightly elongated dimer, in accordance with the
41
42 previously reported value of 2.58 S registered in 40 mM choline, conditions in which the
43
44 protein is fully dimeric 41, whereas a minor percentage (10 %) would be still present as a
45
46
47 somewhat expanded monomer (1.78 S). The sedimentation velocity profile of C-CbpD
48
49 (Figure 4) show in the absence of ligand a major species with sedimentation coefficient
50
51 of 1.44 S and a fractional ratio (f/fo) of 1.27, which is compatible with a globular
52
53
54
monomer. In the presence of choline the major species show a nearly identical
55
56 sedimentation coefficient of 1.46 S, indicating that, contrary to C-LytA and C-Cpl1, the
57
58 monomeric state of C-CbpD does not change upon ligand binding. All these data
59
60
concerning the oligomeric state of the CBMs are summarized in Table 1.
10
1
2
3 Table 1
4
5
6 Kd (µM)
Oligomeric state
7
8 Number of Number of
9 CBRsa binding sitesa SPR
10 + 10 mM Choline DEAE-MNPs
No choline DEAPA chips
choline
11
12
13
14
15 C-Cpl1 670 ± 50 16 ± 4 9±1
Monomer Dimer 12 (6+6)40 5 (2+2+1)
16
17
18
19 C-CbpD 300 ± 10 23 ± 6 N.D.b
20 Monomer Monomer 442 4
21
22 0.19 ± 0.01
23 (25 % signal)c
100 and
24 54 ± 9
Monomer Dimer 1031 (5+5) 833 (4+4) 5300c22
25 C-LytA 14 ± 7
26 (75 % signal)c
27
28
29 LZ-C-LytA 15/10 12/8 N.D.b N.D.b 1.8 ± 0.3
N.D. Trimer/dimer
([5]+5+5) ([4]+4+4)d
30
31
32
33 a
Number of CBRs or sites of the choline-bound species in its corresponding oligomeric state. Parentheses make reference to each of the monomers.
34 b
N.D.: Not determined
35 c
36 High- and low-affinity sites, respectively
d
37 Theoretical estimation
38
39
40
41
42 11
43 ACS Paragon Plus Environment
44
45
46
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51 Figure 4. Sedimentation velocity experiments of CBMs in the absence or presence of choline.
52
53
54
55
56 Estimation of the number of choline binding sites in C-Cpl1 and C-CbpD. As
57
58
59 mentioned above, the crystal structure of the C-LytA module revealed the existence of 4 bound
60
choline residues per monomer of protein.33 On the other hand, no structure of the isolated C-
12
1
2 Cpl1 protein is available, and only the crystal structure of the full Cpl-1 monomer has been
3
4 reported, revealing two choline molecules bound. However, it should be pointed out that such
5
6
structure was elucidated upon soaking the ligand-free protein crystals in a choline solution,37
7
8
9 so that the dimerization process could not take place under such circumstances. Moreover, the
10
11 binding capacity of the isolated C-Cpl1 module, in the absence of the catalytic domain, might
12
13 be different to that within the parental enzyme. Therefore, we decided to experimentally
14
15
16 determine the number of bound ligand molecules for both CBMs by van't Hoff thermal
17
18 analysis.46-47 The thermal stability was assessed by monitoring the circular dichroism (CD)
19
20 signal in the far-UV region (224 nm) at increasing choline concentrations. This technique
21
22
23
reports on both the secondary and tertiary structures of CBMs due to the extensive overlap of
24
25 contributions arising both from the peptide bonds and from the aromatic side chains,22, 31 which
26
27 is reflected in the unusual positive bands in the 220-240 nm region (Figures 5A and B), so that
28
29
this parameter constitutes a suitable probe for the overall structure of these proteins. The ligand
30
31
32 induces an increase in the ellipticity of both proteins, which is more evident for C-Cpl1 (Figures
33
34 5A and B), and confers stability to both proteins by inducing an increase in its denaturation
35
36 temperature midpoint (Tm) (Figures 5C-D). Although the proteins were not able to renature
37
38
39 after cooling down the sample (data not shown), the thermal transitions were fitted to the Gibbs-
40
41 Helmholtz thermodynamic equation (Eq. 1), as we determined that the thermal scans were
42
43 independent of the heating rate (20-50 ºC h-1), so any irreversible step in the denaturation
44
45
46
process is slower than the equilibrium denaturation event.24, 48
47
48
49 The use of the Gibbs-Helmholtz equation requires the previous knowledge, or
50
51 estimation, of the difference in heat capacity between the folded and the unfolded states (ΔCp).
52
53 Several theoretical procedures were followed (see Experimental Section), and the results are
54
55
56 displayed in Table S1. One method is based only on the number of protein residues 49 (Eq. 2).
57
58 Nevertheless, we also made use of the three-dimensional structure of the full-length parental
59
60 Cpl137, from which the structure of the isolated C-Cpl1 module can be deduced to be very
13
1
2
20 20
3 A 0 mM choline B 0 mM choline
4 -3
7 mM choline
15 8 mM choline
5
[] (deg cm dmol ) x 10
10 20 mM choline 20 mM choline
10
6
-1
7 0 5
8 0
2
9 -10 -5
10
11 -10
-20
12 -15
13
-30 -20
14 190 200 210 220 230 240 250 260
190 200 210 220 230 240 250 260 270
15
Wavelength (nm) Wavelength (nm)
16
17
18
C 0 mM choline D 0 mM choline
15
-3
7 mM choline 15 8 mM choline
[]224 (deg cm dmol ) x 10
19 20 mM choline 20 mM choline
20 10
-1
10
21
22
2
5 5
23
24
0
25 0
26
-5
27 -5
28
0 20 40 60 80 0 20 40 60 80
29
30 Temperature (ºC) Temperature (ºC)
31 -3.6
32 E F
33 -4.0 -4.0
34
35
ln [choline]
-4.4
36
37 -4.4
-4.8
38
39
40 -5.2
-4.8
41
42 -5.6
43 3.04 3.06 3.08 3.10 3.06 3.08 3.10
3 -1
44 3
(1/Tm) x 10 (K ) -1
(1/Tm) x 10 (K )
45
46 Figure 5. Circular dichroism studies on C-Cpl1 and C-CbpD. Far-UV CD spectra of C-Cpl1 (A) and C-CbpD
47 (A) in the absence or presence of choline; CD-monitored thermal stability of C-Cpl1 (C) and C-CbpD (D) in
48 the absence or presence of choline; E), van't Hoff plot of C-Cpl1 thermal denaturation; the slope of the fit to
49 Eq. 5 was -21000 ± 2000 K and the mean unfolding enthalpy was 206 ± 20 kcal mol -1, leading to n = 5.0 sites
50 per denaturation unit (dimer); F) van't Hoff plot of C-CbpD thermal denaturation; the slope of the fit to Eq. 5
51 was -22000 ± 1500 K and the mean unfolding enthalpy was 170 ± 20 kcal mol -1, leading to n= 3.9 sites per
52
denaturation unit (monomer). Data shown in panels E and F are the average of at least 3 scans.
53
54
55
56
41
57 similar, as demonstrated by SAXS experiments . Regarding C-CbpD, we made use of the
58
59 homology model described above. With these structures, we employed the theoretical
60
14
1
2 approaches described by Myers et al 50 and Spolar et al 51 (Eqs. 3 and 4, respectively), which
3
4 make use of the differences in accessible surface area between the folded and the unfolded
5
6
states. Despite the theoretical differences among the three approaches, the results of the
7
8
9 calculations were remarkably similar (Table S1), so that an average value of ΔCp was finally
10
11 estimated for C-Cpl1 (2.0 cal mol-1 K-1) and for C-CbpD (1.2 cal mol-1 K-1).
12
13
14 The van´t Hoff analysis of C-Cpl1 thermal unfolding scans (Eq. 5) is shown in Figures
15
16
5E, yielding a value of n = 5.0 sites per cooperative denaturation unit. Since, at the
17
18
19 concentrations tested, C-Cpl1 is a dimer (Figure 4), the results are then compatible with the
20
21 presence of 2 bound ligands per monomer plus an additional one probably located at the
22
23 dimerization interface and configured by residues W210, F218 and Y238 of each monomer, as
24
25
26 previously suggested.41 Regarding C-CbpD, the van't Hoff analysis (Figure 5F) yielded, in this
27
28 case, a value of n = 3.9 ligand molecules per monomer, in agreement with the 4 predicted
29
30 binding sites considered in the structural model of the protein (3 standard sites plus a non-
31
32
33 canonical additional one, Figure 3). It should be pointed out that non-specific effects of choline
34
35 chloride on protein stability due to the increase in ionic strength should be ruled out, at least at
36
37 the concentrations used (0-20 mM), since temperature scans of both C-Cpl1 and C-CbpD either
38
39
40
in the presence or the absence of 20 mM NaCl yielded the same thermodynamical results (data
41
42 not shown).
43
44
45
46
47
48 Affinity binding of soluble choline to CBMs. Choline binding affinity for C-LytA
49
50 protein has been previously determined, unravelling the existence of low- and high-affinity
51
52
53 binding sites (Table 1).22 To estimate the binding constants for C-Cpl1 and C-CbpD,
54
55 equilibrium titrations were carried out by following the change in the intrinsic fluorescence
56
57 spectra upon addition of the ligand. In both cases the addition of choline induces a decrease in
58
59
60
the signal concomitantly with an evident blue-shift (Figure S1), indicating both an enhanced
15
1
2 quenching and a decrease in the polarity of the environment surrounding the tryptophan
3
4 residues. Spectra of the proteins recorded in the presence of 20 mM NaCl showed no significant
5
6
(C-Cpl1) or very slight (C-CbpD) differences in the fluorescence intensity, as well as negligible
7
8
9 changes in the wavelength of peak maximum (Figure S1) therefore discarding important, non-
10
11 specific ionic strength effects. Both C-Cpl1 and C-CbpD display a well-defined hyperbolic
12
13 binding curve when monitoring the absolute value of the changes in average emission
14
15
16 intensities (<>) (Eq. 6, Figure 6). Dissociation constants (Kd) were calculated assuming that
17
18 all binding sites had similar affinity, as previously described for Cpl-1.21 Analysis of titration
19
20
curves (Figure 6) considered the binding of 5 molecules, coupled with dimerization, to C-Cpl1,
21
22
23 assuming a negligible presence of protein dimers in the absence of choline (Eq. 7) or the
24
25 independent binding of 4 molecules to C-CbpD monomers (Eq. 8).
26
27
28
29 4
C-Cpl1 Kd = 670 ± 50 M 5 C-CbpD Kd = 300 ± 10 M
30
31
3 4
32
33
| (nm)
 (nm)
3
34 2
35 2
36
1
37 1
38
39 0 0
40
41 0 5 10 15 20 0 5 10 15 20 25
42 [Choline] (mM) [Choline] (mM)
43
44 Figure 6. Titration curves of C-Cpl1 and C-CbpD with choline, monitored by intrinsic fluorescence (absolute
45 value of changes in average emmision intensity). The average values of triplicates are shown.
46
47
48
49
50 The calculated constants (Kd = 0.67 ± 0.05 mM for C-Cpl1 and 0.30 ± 0.01 mM for C-
51
52 CbpD, Table 1), are intermediate between the high- and low-affinity sites in C-LytA (0.1 mM
53
54 and 5.3 mM, respectively),22 and indicate a higher affinity by C-CbpD for the ligand than C-
55
56
57 Cpl1. It is noteworthy that the Kd value for C-Cpl1 (0.67 mM) is lower than that reported for
58
59 the full-length parental Cpl1 enzyme derived from CD-monitored (3.6 mM) and isothermal
60
16
1
2 titration calorimetry titrations (1.9 mM).21 These discrepancies might arise from the previous
3
4 consideration based on the crystal structure,37 of 4 binding sites instead of 5 that we postulate
5
6
here, and/or from the strong interaction present, in the parental enzyme, between catalytic and
7
8
9 choline-binding modules37 that could somehow obstruct the binding of the ligand and,
10
11 consequently, the dimerization of the protein. In fact, differential scanning calorimetry (DSC)
12
13 studies previously carried out with both Cpl-1 and C-Cpl1 proteins showed that, in the presence
14
15
16 of 10 mM of choline, the differences between the calorimetric and the van´t Hoff denaturation
17
18 enthalpies could be explained if the C-Cpl1 module was present as a dimer whereas only 30 %
19
20 of the parental protein acquired a dimeric conformation in these conditions.24
21
22
23 In any case, the estimated dissociation constants are of high magnitude (millimolar
24
25
26 range), confirming that the affinity of CBMs for free choline is very low in general terms.
27
28 Therefore, to explain the high affinity shown by CBMs for the cell wall, synthetic systems
29
30 should be used that take into account the multivalency effects arising from the multiple
31
32
33 disposition of choline receptor molecules on a supramolecular, polydentate structure.
34
35
36
37
38 Binding of CBMs to solid-state cell-wall mimics. To avoid any discrepancies due to
39
40 the lack of homogeneity of different cell wall preparations that would hamper a suitable
41
42
43 comparative binding analysis using this natural substrate, and taking advantage of the wide
44
45 range of choline analogs able to interact with CBMs52-53 we made use of two types of cell-wall
46
47 mimics functionalized with multiple copies of a choline analogue: i) magnetic nanoparticles
48
49
50
(200 nm) functionalized with 2-diethylaminoethyl-starch groups (DEAE-MNPs), and ii)
51
52 surface plasmon resonance (SPR) chips derivatized with N,N-diethyl-1,3-diaminopropane
53
54 (DEAPA). It should be reminded that DEAE is an usual compound in the affinity
55
56
chromatography resins employed to purify CBPs,52-53 and that the specific binding of CBPs to
57
58
59
60
17
1
2 the DEAE-MNPs has already been previously reported in our laboratory.54 Binding of the three
3
4 CBMs to the DEAE-MNPs was subjected to Langmuir analysis (Eq. 9, Figure 7). The
5
6
calculated apparent Kd values are shown in Table 1. In the case of C-LytA protein, and despite
7
8
9 that protein harbours two types of binding sites for choline22 a single apparent dissociation
10
11 constant could only be discerned (54 ± 9 µM), with other alternative fittings being unsuccessful
12
13 (data not shown). Regarding C-Cpl1 and C-CbpD proteins, the obtained Kd's were comparable
14
15
16 (16 ± 4 and 23 ± 6 µM, respectively) and reflect a higher affinity for the solid substrate than
17
18 C-LytA even though they possess less choline-binding sites (Table 1, Figure 2).
19
20
21 14 12
22 A Kd = 54 ± 9 M B Kd = 16 ± 4 M
(nanomoles/mg DEAE-MNPs)
12
23 10
24 10
8
Bound protein
Bound protein
25
8
26 6
27 6
28 4
4
29
30 2 2
31 0
0
32 0 20 40 60 80 100 120 0 10 20 30 40
33 [C-LytA] (M) [C-Cpl1] (M)
34
35 120
C Kd = 23 ± 6 M
36
100
37
38 80
Bound protein
39
40 60
41
42 40
43
20
44
45 0
46 0 20 40 60 80 100 120
47 [C-CbpD] (M)
48
49
50 Figure 7. Langmuir binding isotherms of CBMs to DEAE-MNPs. Experiments were performed in triplicate.
51
52
53
54 The affinity analysis and binding kinetics of CBMs were also determined by SPR. The
55
56
57 purified proteins were applied to sensor chips coated with DEAPA. Representative
58
59 sensorgrams for C-Cpl1 and C-LytA are presented in Figure 8. The C-CbpD protein could not
60
18
1
2
3 A
4
5
6
7
8
9
10 Kd = 9 ± 1 µM
11
12
13
14
15
16 B
17
18
19
20
Kd(1) = 0.19 ± 0.01 µM
21
Kd(2) = 14 ± 7 µM
22
23
24
25
26 C
27
28
29
30
31
32 Kd = 1.8 ± 0.3 µM
33
34
35
36
37
38 Figure 8. Surface plasmon resonance (SPR) analysis of the interaction of CBMs with DEAPA-coated chips.
39 Left: SPR sensorgrams. Right: Binding plots derived from SPR sensorgrams used to calculate the affinity
40 constants. A) C-Cpl1; B) C-LytA; C) LZ-C-LytA. Experiments were carried out in duplicate, but only a sample
41 is shown in the figure.
42
43
44 be used for SPR experiments due to a non-homogeneous behaviour at any of the densities used,
45
46
probably due to protein aggregation on the surface (data not shown). The binding of C-LytA
47
48
49 to the DEAPA surface (Figure 8B) displayed similar characteristics than those to soluble
50
51 choline, as it could only be analyzed in terms of a two-site affinity model, with 75 % of the
52
53 signal corresponding to the low affinity sites (Kd = 14 ± 7 µM), and 25 % of the signal
54
55
56 representing the high affinity sites (Kd = 0.19 ± 0.01 µM). The binding curve of C-Cpl1 (Figure
57
58 8A) was, conversely, best fitted to a single-site affinity model, (Kd = 9 ± 1 µM), again
59
60 representing an intermediate value between the two types of C-LytA sites (Table 1). The
19
1
2 dissociation constants obtained from the two experiments employing a polydentate, solid
3
4 substrate represent binding affinities 20-500-fold higher than those for the free ligand (Table
5
6
1), highlighting appreciable multivalent effects that may be extrapolated to the bacterial cell
7
8
9 wall.
10
11
12 Effect of exogenously added CBMs to pneumococcal planktonic cultures. The
13
14 results shown so far indicate that CBMs are able to efficiently interact with choline analogs
15
16
contained on the surface of insoluble, multivalent substrates. Our next step was testing whether
17
18
19 CBMs added to a pneumococcal culture could compete with the bacterial CBPs for binding to
20
21 the choline in the cell wall, therefore inhibiting these proteins. In support for this hypothesis, it
22
23 should be mentioned that we had previously described that the C-Cpl1 or C-LytA modules
24
25
26 inhibit the activity of the parental Cpl1 and LytA hydrolases in cell-wall hydrolysis
27
28 experiments in vitro.31 If successful, effects such as prevention of autolysis or inhibition of
29
30 separation of daughter cells upon division might be expected, leading to the formation of long
31
32
33 chains, as it has been long observed upon addition of an excess of choline to the medium.20 To
34
35 facilitate the observations, the experiments were carried out with the R6CIB17 strain, a
36
37 derivative of the commonly used non-capsulated R6 laboratory strain that was identified in our
38
39
40
group and that does not flocculate in liquid medium (see Experimental Section). A detailed
41
42 description of this strain will be published elsewhere. Firstly, we added to a bacterial culture a
43
44 hybrid protein (GFP-C-LytA) in which the green fluorescent protein (GFP) is fused to the C-
45
46 LytA module.55 The protein was added in early exponential phase (OD550 ≈ 0.1) and samples
47
48
49 taken at different times were observed by fluorescence microscopy. As depicted in Figure 9, as
50
51 early as 5 min after addition, the GFP-C-LytA protein was evenly distributed on the bacterial
52
53 surface. As the growth proceeded, long chains were visible and, at the start of the stationary
54
55
56 phase, appreciable cell clumps were detected while most of the fluorescence was kept in the
57
58 bacterial septa (Figure 9). In fact, the septum has been described to be the destination of
59
60
20
1
2 intracellularly expressed, full-length LytA autolysin,56 confirming the preference of the C-
3
4 LytA moiety to the equatorial region in the cell in the later stages of growth.
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51 Figure 9. Fluorescence microscopy micrographs of S. pneumoniae R6CIB17 cultured in the presence of
52 GFP-C-LytA (1 µM). The time after protein addition at an OD 550 of 0.1 is indicated. Arrows indicate the
53 position of bacterial septa.
54
55
56 The pneumococcal strain R6CIB17 was cultured in test tubes without shaking or turning
57
58 over prior to measure the optical density (OD550). In these conditions, cells acquire the usual
59
60 diplococcal shape, or form very short chains (Figure 10, panel i), and finally undergo autolysis
21
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44 Figure 10. Effect of addition of CBMs to pneumococcal planktonic cultures. Left: growth curves. Black arrows
45
indicate time of CBM addition to the medium, and blue arrows indicate time of sample removal for microscopy.
46
47 Experiments were performed in duplicate. Right, confocal fluorescence microscopy micrographs (i-vi) of samples
48 taken from the bottom of the tube in the stationary phase of the culture. Cells were stained with the BacLight kit.
49 Bars indicate 10 µm.
50
51
52
53
54 at the end of the stationary phase, detected as a decrease in optical density in control
55
56 experiments (Figure 10 A-C). Addition of choline, on the other hand, induces the formation of
57
58 long chains (Figure 10, panel ii) that do not autolyse nor flocculate (Figure 10 A-B). To test
59
60
the effect of CBMs, these proteins were added to the culture in the early exponential growth
22
1
2 phase (OD550  0.1) to a final concentration of 40 nM, 400 nM and 1 µM. No effect was shown
3
4 by any of the CBMs tested at 40 nM (data not shown). However, upon addition of 400 nM C-
5
6
7 Cpl1, visible aggregates were soon observed and subsequently deposited in the bottom of the
8
9 tube, leading to a decrease in the OD550 of the supernatant (Figure 10A) (Samples of cell clumps
10
11 observed upon addition of 1 µM CBMs are shown in Figure 10, panels iii-vi). A similar result
12
13
14
was observed with C-CbpD, although in this case visible aggregation was only detected at the
15
16 end of the exponential phase (Figure 10A). Regarding C-LytA, this protein did not lead to
17
18 flocculation at this concentration (400 nM), although a clear prevention of autolysis at the end
19
20
of the stationary phase was evident (Figure 10A). On the other hand, when the concentration
21
22
23 of the CBMs was increased to 1 µM, a much more pronounced effect was achieved, leading to
24
25 cell deposition even in the case of C-LytA (Figures 10 and S2). As mentioned above, no cell
26
27 aggregation was observed when choline was added to the culture (Figures 10A-B and panel ii),
28
29
30 suggesting additional mechanisms of action exerted by the CBM polypeptides compared to the
31
32 aminoalcohol.
33
34
35 Samples of cultures containing 1 µM of each corresponding CBM were taken in the
36
37 stationary phase, both from the upper part and the bottom of the culture tube, and characterized
38
39
40 by confocal microscopy upon staining with the LIVE/Dead BacLight bacterial viability kit. The
41
42 same sample was used to calculate the number of viable cells. Samples of the upper part of the
43
44 tube were almost devoid of cells (data not shown), whereas those from the bottom contained a
45
46
47
vast majority of green-stained cell chains of smaller length than those induced by choline but,
48
49 above all, accumulated as cell aggregates which were of bigger size in the case of C-Cpl1 and
50
51 C-CbpD (Figure 10). Besides, viability tests confirmed that the addition of CBMs to the
52
53
bacterial culture did not induce the cells death (data not shown), in agreement with the
54
55
56 observation that the vast majority of cells were green-labeled in all cases when stained with
57
58 the BacLight kit, demonstrating that the bacterial membrane was not compromised and
59
60 therefore ruling out any lytic effect induced by CBMs (Figure 10). These results suggest that
23
1
2 the bacterial aggregation could arise from the formation of cell chains as a consequence of
3
4 bacterial host CBP inhibition. However, the fact that a chaining phenotype is also achieved
5
6
upon CBP inhibition by addition of choline, but in the latter case the cells are kept in
7
8
9 suspension, suggests that added CBMs must participate in additional cell-cell interactions that
10
11 lead to aggregation. Moreover, those CBMs displaying a higher affinity to DEAE-containing
12
13 supramolecular substrates (C-Cpl1 and C-CbpD, Table 1) were also the most effective in
14
15
16 inducing the flocculation effects.
17
18
19 Enhancing the CBM flocculating effect by protein engineering. As described above,
20
21 the interaction between CBMs and the pneumococcal surface (or mimics thereof) displays the
22
23 classical multivalency features arising from two polydentate entities: a protein with several
24
25
26 choline-binding sites and a substrate with multiple choline/choline analogs. Our results also
27
28 show that there seems to be no direct correlation between the number of choline-binding sites
29
30 and affinity for the substrate (Table 1). Nevertheless, we decided to study the effect of
31
32
33 increasing the number of ligand-binding27 sites per molecule but in a branched, rather than
34
35 linear, disposition, based on dendrimeric structures that have been previously used by our group
36
37 to enhance the effect of choline and choline analogs on pneumococcal cultures.26-27, 57 To do
38
39
40
so, we designed an oligomeric derivative of the C-LytA module (LZ-C-LytA), containing the
41
42 leucine zipper sequence (LZ) of the PhaF protein from Pseudomonas putida KT2440, which
43
44 has been described as a protein oligomerization motif.58-59 The LZ sequence was fused to the
45
46 N-terminus of C-LytA and separated by a linker to prevent possible steric hindrances that
47
48
49 would hamper choline-induced oligomerization (Figure S3). The reason to choose C-LytA was
50
51 to test the real possibilities of our approach to improve the properties of the CBM with the
52
53 smallest affinity for DEAE-containing substrates and the lowest effectiveness on the
54
55
56 pneumococcal cultures.
57
58
59
60
24
1
2 The LZ-C-LytA polypeptide was overexpressed and purified by affinity
3
4 chromatography in DEAE-cellulose as described in the Experimental Section. To determine
5
6
the oligomerization state of the LZ-C-LytA, a sedimentation velocity experiment was carried
7
8
9 out in the presence of 10 mM of choline. The experiment could not be performed in the absence
10
11 of choline due to the aggregation of the sample in the ultracentrifugation conditions. The results
12
13 shown in Figure 4 and Table S2 show that in the presence of choline the protein exists as a
14
15
16 mixture of species: a major one of 2.69 S (64 %), an additional one of 2.20 S (27 %) and two
17
18 minor species of 1.09 S (3 %) and 3.36 S (6 %). The presence of choline complicates the
19
20 analysis to determine the nature of the LZ-C-LytA complexes, since, in these conditions, the
21
22
23
protein could oligomerize via the C-terminal of the C-LytA moiety and/or via the LZ sequence.
24
25 In an attempt to assign each sedimentation signal to the possible oligomeric complexes, we
26
27 generated in silico the three-dimensional structures of all possible oligomers (see Experimental
28
29
Section for details) (Figure S3). Briefly, a model of the LZ sequence58 was linked to C-LytA
30
31
32 to generate the core monomer. Next, the dimer involving interactions between C-LytA C-
33
34 terminal hairpins (CT2 variant) was generated with Swiss-PDB viewer by homology modelling
35
36 based on the crystal structure of the C-LytA dimer,33 while the dimer involving LZ-LZ
37
38
39 interaction (LZ2 variant) was generated using the M-ZDOCK server.60 On the other hand,
40
41 among the variety of possible higher-order oligomers (trimers and tetramers) we decided to
42
43 discard those models containing mixed LZ-LZ and C-LytA-C-LytA interactions. The reason is
44
45
46
that those models would contain unpaired oligomerization motifs that should lead to further
47
48 polymerization, in contrast with the analytical centrifugation results (Figure 4). As a
49
50 consequence, only LZ-based oligomers were considered (LZ3 and LZ4, Figure S3). Next, the
51
52
hydrodynamic properties of the structures were estimated using the HydroPRO utilities.61
53
54
55 Table S2 displays a comparison among the experimental and predicted values of the
56
57 sedimentation coefficients. The major peak in the sedimentation profile (2.69 S) is compatible
58
59 with a trimer, whereas the 2.20 S species could represent a dimer, and the minor peaks might
60
25
1
2 correspond to monomers and tetramers. The theoretical hydrodynamic predictions of the in
3
4 silico models (Table S2) are, within the limits of this approach, reasonably compatible with the
5
6
experimental sedimentation results, prompting us to predict the predominant presence of the
7
8
9 LZ-oligomerized species depicted in Figure S3.
10
11 Next, we determined the affinity of the LZ-C-LytA oligomers to DEAE-functionalized
12
13 surfaces. The binding to DEAE-MNPs could not be carried out due to protein aggregation at
14
15
16 the concentrations needed. However, SPR experiments, which make use of a much lower
17
18 concentration, yielded the sensorgram shown in Figure 8C. Unlike C-LytA (Figure 8B), data
19
20 for LZ-C-LytA could only be properly fitted assuming one type of binding sites (Kd = 1.8 ±
21
22
23
0.3 µM, Table 1). The high-affinity sites of C-LytA are involved in protein dimerization
24
25 through the C-terminal hairpins.23, 34, 62 However, this interaction might be hampered by the
26
27 alternative LZ-LZ-mediated oligomerization (Figure S3), which could leave the C-terminal
28
29
hairpins withouth the possibility of interaction, mostly due to steric hindrance reasons, and with
30
31
32 the result of the conversion of all binding sites to the "low-affinity" kind. In support of this, it
33
34 has been previously described that LytA variants which had their C-terminal tags truncated or
35
36 deleted do not display high-affinity binding sites anymore.62-63 With this hypothesis in mind,
37
38
39 upon comparison with the affinity of the major fraction of C-LytA sites for SPR chips (75% of
40
41 population, Kd =14 µM), the incorporation of the LZ motif would result in an 8-fold increase
42
43 on the affinity. This would indicate that the oligomerization of a protein (C-LytA) with
44
45
46
moderate binding capacity leads to a significant increment in the affinity for its ligand. In fact,
47
48 LZ-C-LytA turned out to be the best binder among those CBMs tested by SPR (Table 1).
49
50 To assess whether the enhancement in binding properties shown by LZ-C-LytA had a
51
52
reflection on the bacterial culture assays the protein was added to a planktonic culture of
53
54
55 R6CIB17 at early exponential phase (Figure 10C). A slight effect on the OD550 was observed
56
57 at a concentration as low as 40 nM of added LZ-C-LytA, similar to that obtained at 400 nM of
58
59 C-LytA (Figure 10A). Nevertheless, results are more evident at 100 nM concentration which
60
26
1
2 caused eye-visible bacterial aggregation and a OD decrease comparable to that obtained using
3
4 higher concentrations of C-Cpl1 (400 nM) and C-LytA (1 µM) (Figure 10C). Moreover, the
5
6
confocal images of BacLight-stained samples taken from the bottom of the tube containing 0.1
7
8
9 µM of LZ-C-LytA (stationary phase) showed the accumulation of long, green-stained cell
10
11 chains (Figure 10, panel vi), whereas the number of viable cells remained unchanged (data not
12
13 shown). These results indicate that the multivalent presentation of CBMs in a branched
14
15
16 disposition exponentially augments the binding affinity to the choline in the cell wall and
17
18 concomitantly reduces the necessary concentrations to induce bacterial aggregation.
19
20
21
22
23
Phagocytosis of S. pneumoniae previously treated with CBMs by peritoneal
24
25 macrophages. Our results suggest that the incubation of pneumococcal cultures with a CBM
26
27 does not arrest the cell division or decrease viability but induce the cell aggregation and/or
28
29
establishment of long chains, and these effects could in fact facilitate bacterial phagocytosis by
30
31
32 the host macrophages, as we have previously demonstrated for cultures treated with choline
33
34 dendrimers.27 To test this hypothesis, we compared the number of intracellular colony forming
35
36 units (CFU) after 2 and 16 h of pre-treatment of a R6 pneumococcal culture either with C-Cpl1
37
38
39 or LZ-C-LytA at a final concentration of 1 µM with the number of bacteria phagocytosed by
40
41 macrophages without CBM pre-treatment (please note that here we used the R6 naturally
42
43 flocculating strain and not our R6CIB17 derivative used in the planktonic assays). To account
44
45
46
for intraday variation of the assay, the numbers of bacteria phagocytosed after CBM pre-
47
48 treatment within 1 h was expressed in % of the number of bacteria phagocytosed in the same
49
50 time interval without pre-treatment (Figure 11). Compared to the control samples without CBM
51
52
added, the phagocytosis of bacteria substantially increased when the culture was previously
53
54
55 treated with the CBMs, irrespective of the co-incubation time (Figure 11). Remarkably, the
56
57 percentage of intracellular CFU was nearly 3 times higher with LZ-C-LytA compared to C-
58
59 Cpl1, mirroring their binding affinities to SPR DEAPA chips in vitro (5-fold higher affinity for
60
27
1
2 the former protein, Table 1 and Figure 8) and the agglutination effects in planktonic cultures
3
4 (Figure 10), where a 0.1 µM concentration of LZ-C-LytA produced approximately the same
5
6
aggregation effect (as measured by optical density) as a 4-5 fold higher concentration of C-
7
8
9 Cpl1 (0.4 µM). In any case, when carrying out these comparisons, it should be taken into
10
11 account that the aspects of the aggregates produced by the two proteins are somewhat different,
12
13 with longer chains in the case of LZ-C-LytA (Figure 10), and this may have some influence
14
15
16 both on the sedimentation properties of the cell aggregates in the liquid cultures and, above all,
17
18 on a highly complex phenomenon such as phagocytosis.
19
20
21
22
23
24 ****
2500
25 ****
re fe rre d to th e c o n tro l
26
in t r a c e llu la r C F U %
2000 ****
27 ****
28 1500
29 1000
30
31 500
32
33
0
c o n tro l 2h 16h c o n tro l 2h 16h
34
C - C p l- 1 L Z -C L y tA
35
36
37 c o - in c u b a tio n o f c e lls
c o - in c u b a tio n o f c e lls
38 a n d b a c te r ia
a n d b a c te r ia
39 a d d itio n o f L Z - C L y tA a d d itio n o f C - C p l1
40 10
10
41
c o n tr o l
c o n tr o l
8 C - C p l1 2 h
42 8 L Z - C L y tA 2 h
lo g 1 0 C F U /m l
C - C p l1 1 6 h
lo g 1 0 C F U /m l
L Z -C L y tA 1 6 h
43 6
6
44 4
4
45
46
2
2
47 0 0
48 0 2 4 6 8 10 12 14 16 18 20 22 0 2 4 6 8 10 12 14 16 18 20 22
t im e ( h ) t im e ( h )
49
50
51 Figure 11. Pre-treatment of pneumococci with choline-binding modules (CBM) increases their phagocytosis
52 by macrophages. Comparison of the numbers of intracellular colony forming units (CFU) after 2 h (red bars)
53 and 16 h (green bars) of pre-treatment of a R6 pneumococcal culture either with 1 µM C-Cpl1 (cells from 7
54 animals; 20-21 measurements/group) or 1 µM LZ-C-LytA (cells from 5 animals; 19-20 measurements/group).
55 The numbers of bacteria phagocytosed after CBM pre-treatment within 1h were expressed in % of the numbers
56 of untreated bacteria phagocytosed in the same time interval. Data are expressed as medians (25th
57 percentile/75th percentile). Kruskal-Wallis test followed by Dunn’s multiple comparison test was used to
58 compare data (****, p < 0.0001).
59
60
28
1
2 DISCUSSION
3
4 The choline-binding protein system of S. pneumoniae constitutes a paradigmatic
5
6
example of modular evolution40, 64 that is essential for pneumococcal viability and virulence.18
7
8
9 This reason, together with the common mechanism shared by all members of the family to
10
11 recognize the choline residues contained in the cell-wall, shows clues for the development of
12
13 compounds that might simultaneously inhibit or interfere with their activity and that could
14
15
16 represent a novel set of antimicrobials to tackle current antimicrobial resistance. Since the use
17
18 of lytic compounds may not be devoid of harmful consequences due the uncontrolled release
19
20 of virulence factors and toxic compounds from the bacterial cytoplasm,28-29 we decided to
21
22
23
investigate the effect of exogenous CBMs as surface choline blockers, with the aim of
24
25 generating cell chains and aggregates that reduce bacterial dissemination, promote host
26
27 phagocytosis and, eventually, favour complement deposition as previously described 32 (Figure
28
29
1).
30
31
32 Initially we carried out a detailed structural analysis of three representative CBMs,
33
34 namely C-LytA, C-Cpl1 and C-CbpD, differing in size, number of CBRs and binding sites, as
35
36 well as in oligomeric state (Figures 2-5, Figure S1 and Table 1). Their affinity for free choline
37
38
39 and polydentate supramolecular substrates was also assessed (Figures 6-8 and Table 1). Then,
40
41 taking advantage of their ability to bind to this cell wall mimics, they were assayed on
42
43 pneumococcal planktonic cultures (Figures 9-10 and S2) with the aim of saturating the choline
44
45
46
residues in the teichoic and lipoteichoic acids and make them unavailable for the full-length
47
48 CBPs. Remarkably, in addition to the formation of long chains as a consequence of inhibiting
49
50 those CBP hydrolases involved in separation of daughter cells upon division (similar to the
51
52
addition of choline), the three tested CBMs also induced, to a variable extent, the aggregation
53
54
55 of pneumococcal cells, even leading to visible clumps (Figures 9-10 and S2). A rationally
56
57 engineered form of a CBM (Figure S3) successfully made use of multivalency principles to
58
59 enhance its substrate recognition and aggregating activity (Figures 8 and 10, Table 1). Finally,
60
29
1
2 these observations were exploited to check the ability of the CBMs to augment bacterial
3
4 phagocytosis by mouse peritoneal macrophages (Figure 11). All these results together point to
5
6
CBMs as a novel streamlined alternative of non-lytic antipneumococcal compounds aimed to
7
8
9 facilitate the natural defense mechanisms of the infected host.
10
11 Many intriguing questions have arisen from the pneumococcal CBP system from the
12
13 structural point of view given the wide variety in geometries and number and disposition of
14
15
16 CBRs displayed by CBMs.18 For instance, is the affinity for choline directly related to the
17
18 number of choline-binding sites? Previous studies suggested that this is not necessarily the
19
20 case, at least regarding free choline. The affinity for free choline in solution has been shown to
21
22
23
be similar both for LytC and Cpl1,21 but the former contains 7 binding sites per monomer38
24
25 while the latter possesses 5 sites per dimer, as we show here. In the same line, full-length LytA
26
27 contains 10 canonical and 2 non-canonical sites per dimer,39 but it apparently displays a lower
28
29
ligand affinity than Cpl1, with only 5 sites per dimer.21-22 Our results obtained with the CBMs
30
31
32 confirm these observations as the affinity of C-CbpD for choline (4 binding sites per monomer,
33
34 Figures 3 and 5) is even higher than that of C-Cpl1 (5 binding sites per dimer, Figure 5).
35
36 Nevertheless, it should be taken into account that the actual substrate of CBMs is the cell wall,
37
38
39 which contains multiple copies of choline distributed along this supramolecular structure,
40
41 leading to multivalency effects. So far, no affinity studies of CBPs have taken this into
42
43 consideration, most likely due to the inherent inhomogeneities in different cell wall
44
45
46
preparations. Therefore, we decided, for the first time, to use two defined, supramolecular cell-
47
48 wall mimics (DEAE-MNPs and SPR DEAPA chips) to evaluate the binding strength of CBMs.
49
50 Table 1 shows that the binding affinity to these substrates increases by a factor of 20-500
51
52
compared to free choline, dissociation constants dropping from the millimolar to the
53
54
55 micromolar range (Figures 7 and 8). The multivalent surface display is crucial for many
56
57 biological interactions.65 Its effect is widely recognized as an important strategy to improve
58
59 affinity and specificity for biological targets.66-67 In fact, we have previously proven that
60
30
1
2 dendrimeric cell-wall mimics containing 4-64 choline molecules per particle display an affinity
3
4 to CBMs several orders of magnitude higher than monovalent choline.26 The comparative study
5
6
among the tested CBMs show that both C-Cpl1 and C-CbpD have a greater affinity to the
7
8
9 macromolecular substrates than C-LytA, despite this polypeptide harbouring more choline-
10
11 binding sites (Table 1). This is in agreement with our previous observation that C-Cpl1 is also
12
13 a more efficient inhibitor in vitro of the activity of parental LytA and Cpl1 hydrolases.31 On
14
15
16 the other hand, the engineered LZ-C-LytA oligomer, displaying the CBRs in a branched, rather
17
18 than linear, fashion (Figure S3), also demonstrated an enhancement in affinity to SPR matrices
19
20 compared to C-LytA alone (Figure 8 and Table 1). All these results strongly suggest that,
21
22
23
despite the high sequence similarity displayed among most CBRs described so far, the
24
25 biophysical basis of choline recognition in the cell wall by CBPs is a complex phenomenon
26
27 which deserves further structural studies.
28
29
The affinity data discussed above, overall, can be translated to the effect of CBMs on
30
31
32 pneumococcal planktonic cultures: both C-Cpl1 and C-CbpD caused an appreciable inhibition
33
34 of CBP action (formation of long chains) and subsequent cell clumping at lower concentrations
35
36 than the C-LytA protein (Figure 10), but the effect was most evident with LZ-C-LytA, which
37
38
39 also displayed the strongest binding to SPR DEAPA chips (Figure 8 and Table 1), as well as
40
41 the strongest stimulation of phagocytosis by macrophages (Figure 11). These results validate
42
43 the use of DEAE-MNPs and DEAPA chips as suitable, defined in vitro models of the
44
45
46
pneumococcal cell wall regarding choline disposition, which may be employed for the design
47
48 and evaluation of novel variants.
49
50 Induction of cell aggregates suggests a second role of isolated CBMs as non-covalent
51
52
agglutinating agents between the cells, leading to sedimentation. This is likely to be due to a
53
54
55 thorough coating of the cell envelope by the protein added, increasing the hydrophobicity of
56
57 the surface. In fact, a detailed inspection of the surface properties of the three tested CBMs
58
59 (Figure 2B) reveal a certain degree of amphipathicity in their structures, with the presence of
60
31
1
2 extensive hydrophobic patches that may lead to non-specific protein-protein associations
3
4 among adsorbed CBMs. Alternatively, due the solenoid shape of CBMs that allows the
5
6
distribution of choline-binding sites all around the peptide chain (Figure 2A), there is also the
7
8
9 possibility that a single CBM molecule binds to choline residues provided by two different
10
11 cells, thus leading to flocculation. These possibilities are summarized in Figure S4. These cell
12
13 clusters were more readily phagocytosed by mouse peritoneal macrophages than untreated
14
15
16 pneumococci (Figure 11). This points to the CBMs as a possible novel family of antimicrobial
17
18 agents for the management of pneumococcal infection by activating the natural host defense
19
20 response (by inducing the phagocytosis of cell aggregates) as well as presumably blocking
21
22
23
bacterial dissemination and favouring complement deposition (by promoting cell chains)32 and
24
25 preventing adhesion (for instance by displacing the CbpA adhesin), without any of the possible
26
27 drawbacks derived from non-controlled cellular lysis. In this sense, it has been demonstrated
28
29
that agglutinating antibodies are more effective than nonagglutinating ones at promoting
30
31
32 bacterial clearance in systemic infection models,68 and much earlier observations reported that
33
34 administration of pneumococci to immunized rabbits caused an almost immediate bacterial
35
36 aggregation and cleareance.69Therefore, CBMs could represent a type of antibody-like
37
38
39 molecules specifically aimed to the pneumococcal surface that might also be used in
40
41 biomedical applications such as diagnosis and directed drug delivery. Besides, full-length CBP
42
43 enzymes have been successfully assayed in multiple occasions as efficient antipneumococcal
44
45
46
molecules in animal models,18 demonstrating that their choline-binding modules must be stable
47
48 and non-toxic at least in the tested conditions. Taking into account that the CBMs employed in
49
50 this work are natural modules (with the exception of LZ-C-LytA) not optimized for a strong
51
52
binding to the cell wall, these polypeptides are susceptible of been further improved in their
53
54
55 stabilities and binding affinities by protein engineering procedures, so as to became a
56
57 therapeutically attractive antipneumococcal alternative.
58
59
60
32
1
2 CONCLUSIONS
3
4 The affinity of pneumococcal choline-binding modules (C-Cpl1, C-CbpD, C-LytA) for
5
6
choline or pneumococcal cell-wall mimics is not directly related to the number of binding sites
7
8
9 or their native oligomeric state. Theoretical studies and thermodynamic characterization of the
10
11 protein-ligand interaction support the existence of non-canonical binding sites both in C-Cpl1
12
13 and C-CbpD. Moreover, the affinities of these modules for their macromolecular substrates
14
15
16 may be substantially enhanced by rationally engineering their oligomerization (LZ-C-LytA
17
18 protein) thanks to multivalence effects. These in vitro results have been extrapolated to a
19
20 biological environment, as CBMs exogenously added to planktonic cultures recognize, in
21
22
23
nanomolar concentrations, the choline residues in the pneumococcal cell wall, displacing the
24
25 choline-binding proteins of the bacterial host and inducing chain formation and cell aggregates
26
27 that foster their phagocytosis by macrophages. Both cell-clumping and phagocytosis-
28
29
enhancing activities are closely related to the binding strength of CBMs to their
30
31
32 macromolecular substrates in vitro. CBMs constitute, therefore, a suitable system to recognize
33
34 and immobilize pneumococcal cells and represent a potential alternative to combat
35
36 antimicrobial resistance issues by activating the natural defenses while avoiding any possible
37
38
39 negative effects of bacterial lysis.
40
41
42
43
44
45 METHODS Reagents. Choline chloride and DEAE-cellulose were obtained from Sigma-
46
47 Aldrich. Owing to the hygroscopic properties of choline, concentrated stock solutions were
48
49
50
prepared from a freshly opened bottle and stored in aliquots at -20 ºC. Isopropyl β-D-1-
51
52 thiogalactopyranoside (IPTG) was purchased from Apollo Scientific (UK). Magnetic
53
54 nanoparticles (MNPs) functionalized with N, N-diethylethanolamine (DEAE)-starch (200 nm)
55
56
were purchased from Chemicell.
57
58
59
60
33
1
2 Bacterial strains and growth conditions. The Escherichia coli strains were RB79170
3
4 and BL21(DE3) (Novagen) and the cultures were grown with aeration in LB medium or in LB
5
6
plus ampicillin (100 µg/mL) when required.71
7
8
9 The S. pneumoniae strains used in this work were R6, a non-encapsulated strain derived
10
11 from the capsular type 2 clinical isolate strain D39,72 and R6CIB17, a R6 spontaneous mutant
12
13 strain with a non-flocculant phenotype isolated in the laboratory of Profs. Ernesto García and
14
15
16 Pedro García (CIB, CSIC, Spain), and whose whole-genome sequence was determined by
17
18 Macrogen (Seoul, Korea) using a HiSeq sequencer and the TruSeq Nano DNA Kit (Illumina)
19
20 (GenBank accession number: CP038808). Pneumococcal liquid cultures were grown at 37 ºC
21
22
23
without aeration in C medium supplemented with 0.08 % (w⁄v) yeast extract (C+Y medium).73
24
25 Growth was monitored by measuring the optical density at 550 nm (OD550) in an Evolution
26
27 201 spectrophotometer (Thermo Scientific). Prior to every measurement, tubes were normally
28
29
turned over several times to homogenize the solution, except when the flocculating capacity of
30
31
32 additives was being assayed.
33
34 The number of viable cells was determined by counting the colonies appeared from
35
36 appropriate dilutions of culture (in triplicate) after overnight incubation at 37 °C on trypticase
37
38
39 soy plates (Conda-Pronadisa) supplemented with 5 % defibrinated sheep blood (Thermo-
40
41 Fisher).
42
43 Three-dimensional modeling of C-CbpD. A tertiary structure model of C-CbpD was
44
45 43
46
generated by the SwissModel utilities (https://swissmodel.expasy.org/). The best template
47
48 encompassed the residues 234-331 of the pneumococcal CbpJ protein (PDB code 6JYX). 44
49
50 In silico docking of choline molecules was accomplished as follows: a PDB file of
51
52
choline was created with the ChemOffice 10.0 utilities (CambridgeSoft, U.K.) and, together
53
54
55 with the template, prepared for docking using the UCSF Chimera 1.10 package (Resource for
56
57 Biocomputing, Visualization, and Informatics at the University of California, San Francisco -
58
59 supported by NIGMS P41-GM103311). 74 Both target and ligand files were submitted to the
60
34
1
2 SwissDock server (http://www.swissdock.ch/).45 Conformations showing occupancy of the
3
4 four presumed binding sites and with the lowest calculated energy, were selected, and their
5
6
coordinates added to those of the template into a single PDB file.
7
8
9 Expression and purification of proteins. C-LytA and C-Cpl1 proteins were
10
11 overexpressed in E. coli RB791 from plasmids pCE17 and pCM1 plasmids as previously
12
13 described.31 Recombinant plasmids for the overproduction of C-CbpD (pET-ccbpD) and LZ-
14
15
16 C-LytA (pET-lzcclytA) were synthetized by GenScript (New Jersey, USA). For C-CbpD, a
17
18 codon-optimized synthetic nucleotide sequence corresponding to amino acids from 353 to 448
19
20 of parental CbpD protein from the R6 strain was inserted between the NcoI and XhoI restriction
21
22
23
sites of plasmid pET21d(+) (Novagen). Regarding LZ-C-LytA, a codon-optimized synthetic
24
25 nucleotide sequence corresponding to the 106-142 region belonging to the PhaF phasin from
26
27 Pseudomonas putida KT2440, which contains a leucine-zipper motif involved in
28
29
oligomerization58-59 was fused upstream the coding sequence for the C-LytA module as that
30
31
32 expressed from pCE17 plasmid,31 and subsequently cloned into the pET21d(+) vector between
33
34 the NcoI and XhoI restriction sites. In both constructions, three termination codons were
35
36 introduced just before the XhoI site. For the overproduction of C-CbpD and LZ-C-LytA, E.coli
37
38
39 BL21(DE3) cells harbouring either the pET-ccbpD or pET-lzcclytA plasmid were cultured at
40
41 37 ºC in LB medium to an OD600 of 0.6. Expression was then induced by the addition of 0.5
42
43 mM IPTG for 3 h at 30 ºC, and then cells were harvested and disrupted using a Branson sonifier.
44
45
46
Purification of all CBMs from crude extracts was accomplished by affinity
47
48 chromatography on DEAE-cellulose as described before.31, 52 The proteins were selectively
49
50 eluted from the DEAE-cellulose column with 20 mM sodium phosphate buffer pH 7.0 plus 143
51
52
mM choline, dialysed at 4 oC against 20 mM sodium phosphate buffer pH 7.0 to remove the
53
54
55 ligand used for elution and stored at -20 oC. In the case of LZ-C-LytA, to prevent protein
56
57 precipitation, dialysis of the protein was performed at a maximum concentration of 8 µM
58
59 against phosphate buffer with gradually decreasing choline concentrations down to 10 mM,
60
35
1
2 since the protein was highly insoluble in the absence of choline. For SPR experiments, a diluted
3
4 sample of LZ-C-LytA (maximum 4 µM) was dialyzed against 10 mM ammonium
5
6
hydrogencarbonate, centrifuged and subsequently lyophilized.
7
8
9 Protein concentration was determined spectrophotometrically (Evolution 201, Thermo
10
11 Scientific) using a molar extinction coefficient at 280 nm (280) of 62540, 58900, 81076 and
12
13
14
67380 M−1 cm−1 for C-LytA, C-CbpD, C-Cpl1 and LZ-C-LytA respectively.
15
16 Analytical ultracentrifugation. Sedimentation velocity experiments were carried out
17
18 by spinning the protein solution (0.85-5.5 µM) in a 20 mM sodium phosphate buffer, pH 7.0
19
20
in the absence or in the presence of 10 mM choline at 48,000 rpm and 20 ºC in an XLI
21
22
23 analytical ultracentrifuge (Beckman-Coulter Inc.) with UV-VIS optical detection system, using
24
25 an An50Ti rotor and 12 mm double-sector centrepieces. Sedimentation profiles were registered
26
27 every 5 min at 230 nm in case of C-Cpl1 and C-CbpD proteins and at 280 nm in case of LZ-C-
28
29
30 LytA. The sedimentation coefficient distributions were calculated by least-squares boundary
31
32 modelling of sedimentation velocity data using the c(s) method, as implemented in the SEDFIT
33
34 program.75 These s-values were corrected to standard conditions (water, 20 ºC, and infinite
35
36
37
dilution)76 using the SEDNTERP program to get the corresponding standard s-values (s20,w).77
38
39 Circular dichroism spectroscopy (CD). CD-monitored thermal denaturation
40
41 experiments were carried out in a Jasco J-815 spectropolarimeter (Tokyo, Japan) equipped with
42
43
a Peltier PTC-423S and using a cuvette path-length of 0.1 cm. All measurements were carried
44
45
46 out in quadruplicate at 20 ºC in the presence 20 mM sodium phosphate buffer at pH 7.0. Protein
47
48 concentration was 0.1 mg mL-1. The sample was layered with mineral oil to avoid evaporation,
49
50 and the heating rate was 60 ºC h-1. After subtracting the baseline from the sample spectra, CD
51
52
53 data were processed with the adaptive smoothing method provided by the Jasco Spectra
54
55 Analysis software. Molar ellipticities ([θ]) were expressed in deg cm2 (dmol of residues)-1.
56
57 The thermal scans were fitted by least squares to the Gibbs-Helmholtz equation (Eq. 1):
58
59
60
36
1
2 ∆𝐺 𝑜 (𝑇) = ∆𝐻𝑚 (1 − 𝑇⁄𝑇 ) − ∆𝐶𝑝 [(𝑇𝑚 − 𝑇) + 𝑇𝑙𝑛 (𝑇⁄𝑇 )] [Eq. 1]
3 𝑚 𝑚
4
5
6
7 where ΔGo (T) is the free energy of the protein at a temperature T, ΔHm is the van’t Hoff
8
9
10
enthalpy, Tm is the midpoint of denaturation (in Kelvin) and ΔCp is the difference in heat
11
12 capacity between the native and denatured states.
13
14 For ΔCp estimation, several procedures were followed. The rule-of-thumb approach by
15
16
Pace (1989)49 is the simplest one and assigns an average value per residue (Eq. 2):
17
18
19
20
21 ∆𝐶𝑝 = 12 𝑛 [Eq. 2]
22
23
24
25
26 where the units of ΔCp are kcal mol-1 K-1 and n is the number of residues.
27
28 The methods by Spolar et al (Eq. 3)51 and Myers et al (Eq. 4)50 make use of differences
29
30
both in non-polar (ASAnp) and polar (ASAp) accessible surface area between the folded and
31
32
33 the unfolded states:
34
35
36
37
∆𝐶𝑝 = 0.32(∆𝐴𝑆𝐴𝑛𝑝 ) − 0.14(∆𝐴𝑆𝐴𝑝 ) [Eq. 3]
38
39
40 ∆𝐶𝑝 = 0.28(∆𝐴𝑆𝐴𝑛𝑝 ) − 0.09(∆𝐴𝑆𝐴𝑝 ) [Eq. 4]
41
42
43
44
45 where ASA is given in nm2 and the units of ΔCp are kcal mol-1 K-1. In turn, to estimate ASA
46
47 values, in a first step, three-dimensional theoretical structures of extended unfolded states were
48
49 generated with SwissPDB viewer 4.1.0 78 by assigning β-structure Ramachandran angles to all
50
51
52 residues. Then, the structures of both the folded and the unfolded conformations were
53
54 submitted to the GetArea server (http://curie.utmb.edu/getarea.html). 79
55
56 To ascertain the number of choline-binding sites for C-Cpl1 and C-CbpD proteins,
57
58
59 thermal scans were performed at different saturating choline concentrations (see Figure 5). The
60
number of sites was obtained from van´t Hoff plots, making use of the van´t Hoff equation:47
37
1
2 ∆𝐻𝑚⁄ 1 ∆𝑆𝑚⁄
𝑙𝑛([𝐶]) = − ( 𝑛𝑅 ) ∗ ⁄𝑇𝑚 + 𝑛𝑅 [Eq. 5]
3
4
5 where [C] is the molar choline concentration, Tm is the midpoint of denaturation (in Kelvin),
6
7 ΔHm and ΔSm are the average van’t Hoff enthalpy (as calculated with Eq. 1 for each point) and
8
9
10 entropy changes at the midpoint, respectively, R is the gas constant and n is the number of sites.
11
12
13
14 Fluorescence spectroscopy. Fluorescence measurements were performed on a PTI-
15
16
17
fluorimeter TI-Quanta Maser QM-62003SE (Birmingham, NJ, EE.UU) using a 5×5 mm path-
18
19 length cuvette and a protein concentration of 1 µM in 20 mM sodium phosphate buffer at pH
20
21 7.0 and 20 ºC. For choline titrations, aliquots from a 150 mM choline stock solution were added
22
23 stepwise and incubated approximately for 5 min prior to measurement. Tryptophan emission
24
25
26 spectra were obtained using an excitation wavelength of 280 nm and the emission was
27
28 registered between 300 and 400 nm, using excitation and emission slits of 0.5 nm and a scan
29
30 rate of 60 nm·min−1.
31
32
33 The average emission intensity (⟨𝜆⟩) of fluorescence spectra was calculated as depicted
34
35 in Eq .6:
36
37 𝑛 (𝐼
⟨𝜆⟩ = ∑1 𝑖 𝜆𝑖 )⁄ [Eq. 6]
38 ∑1𝑛 𝐼𝑖
39
40
41
42
43 where Ii is the fluorescence intensity measured at wavelength λi .
44
45 The choline titration data for C-Cpl1 were fitted to Eq. 7, which is a modified version
46
47
of that developed by Monterroso et al,21 but this time considering both dimer formation on
48
49
50 choline binding and the presence of 5 choline-binding sites per dimer (see text for details):
51
52
53
54 |∆⟨𝜆⟩| = ([𝑀]2 𝑟 (𝐴 + 𝐵) − 𝑝[𝐷]0 )/[𝐶] [Eq. 7]
55
56
57
58
59 in which the value of [M], A and B is described as follows:
60
38
1
1⁄
2 (−1 + (1 + 8[𝐶]𝑟(1 + 𝐾𝑏 [𝐿])5 ) 2)
3 [𝑀] =
4 (4𝑟(1 + 𝐾𝑏 [𝐿])5 )
5
6 𝐴 = 𝑝 (1 + 𝐾𝑏 [𝐿])5
7
8
9 𝐵 = 5ℎ(𝐾𝑏 [𝐿] + 4𝐾𝑏 2 [𝐿]2 + 6𝐾𝑏 3 [𝐿]3 + 4𝐾𝑏 4 [𝐿]4 + 𝐾𝑏 5 [𝐿]5 )
10
11
12
13 where [M] is the concentration of the monomeric conformation of the protein; [D]0 is
14
15
16 the initial concentration dimers in the absence of choline (assumed in our case to be negligible
17
18 - see Figure 4); [C] is the total protein concentration (in monomer moles); r is the dimerization
19
20 constant in the absence of choline; p is the change in  due to the dimerization of the protein
21
22
23 in the absence of choline; Kb is the choline binding constant and h is the change in  due to
24
25 binding of choline to the protein dimer (per mol of sites in each monomer). The choline
26
27
dissociation constant (Kd) was then calculated as 1/Kb.
28
29
30 The choline titration data for C-CbpD were fitted to a simple ligand binding model (Eq.
31
32 8), considering 4 independent binding sites of the same type:
33
34
35
36
37 4|〈〉|𝑚𝑎𝑥 [𝐿]
|∆⟨𝜆⟩| = [Eq. 8]
38 𝐾𝑑 + [𝐿]
39
40
41
42
43 where  is the absolute change in average emission intensity at each ligand concentration,
44
45  max is the maximum change in average emission intensity per site on ligand saturation, and
46
47
48 Kd is the dissociation constant.
49
50 Binding of CBMs to DEAE-coated magnetic nanoparticles. FluidMAG magnetic
51
52 nanoparticles functionalized with 2-diethylaminoethyl-starch groups (DEAE-MNPs) (200 nm
53
54
55
diameter, Chemicell, Germany) were washed, equilibrated in 20 mM sodium phosphate, pH
56
57 7.0 and finally distributed in aliquots. Such aliquots were then incubated separately with 300
58
59 μL of 20 mM sodium phosphate, pH 7.0 containing increasing concentrations of C-LytA, C-
60
39
1
2 CbpD or C-Cpl1, to a final nanoparticle concentration of 0.25 mg mL-1 (approx. 5.5 × 1010
3
4 particles mL-1). The incubation was maintained for 30 min at room temperature in a rotary
5
6
mixer (Ovan). At this point, bound protein was separated from the solution using a magnet
7
8
9 (MagnetoPURE) for 5 min. The supernatant was subsequently centrifuged 3 min at 5000 × g
10
11 in a miniSpin centrifuge (Eppendorf) to ensure the removal of residual MNPs, and the amount
12
13 of unbound protein was measured spectrophotometrically. The amount of bound protein at each
14
15
16 point was estimated by subtraction of the quantified unbound protein from the initial protein
17
18 added in the assay. The protein binding on the surface of DEAE-MNP was analysed by
19
20 Langmuir analysis80 (Eq. 9):
21
22
23
24
25 𝑞𝑚𝑎𝑥 · [𝐶]
26 𝑞= [Eq. 9]
𝐾𝑑 + [𝐶]
27
28
29
30
31 where q is the amount of protein bound to the MNP at each protein concentration, qmax is the
32
33 maximum adsorption capacity, [C] is the initial protein concentration and Kd is the dissociation
34
35
constant. Non-functionalized MNPs were assayed as controls for non-specific binding, which
36
37
38 was found to be negligible (data not shown). More details on the biophysical characterization
39
40 of the specific binding of the CBMs to DEAE-MNPs will be published elsewhere.
41
42
43
44
45 Surface Plasmon Resonance experiments (SPR). SPR experiments were performed
46
47 at 25 °C with a Biacore X-100 (GE) equipment. Carboxymethyldextran chips (CM5) (Biacore,
48
49 GE) were used as support for immobilization of 3-diethylaminopropylamine. First,
50
51
52
carboxymethylated groups were activated by injection of a 1:1 mixture of 0.4 M 1-ethyl-3-(3-
53
54 dimethylaminopropyl)-carbodiimide and 0.1 M N-hydroxysuccinimide at a flow rate of 10 μL
55
56 min-1 for 7 min. Then, 3-diethylaminopropylamine was injected at 10 μL min-1, using
57
58
concentrations of 1-100 μM for 100 s to obtaining a response density of 250 response units.
59
60
40
1
2 Finally, blockage of the remaining reactive groups was carried out by injection of 1 M
3
4 ethanolamine pH 8.5 for 7 min (flow cell without ligand was used as reference).
5
6
Binding of CBMs in 20 mM sodium phosphate, pH 7.0 plus 0.005 % Tween was
7
8
9 measured by injection of increasing concentrations of each protein over the functionalized chip.
10
11 The flow rate (Vf), the association and dissociation period (tas and tdis) for the different proteins
12
13 tested is indicated in Table S3.
14
15
16 The binding affinity was determined by fitting the sensorgram curve obtained to a single
17
18 hyperbole for LZ-C-LytA and C-Cpl1 (model with one type of CBSs, Eq. 10) and to a double
19
20 hyperbole for C-LytA (model with two types of CBSs, Eq. 11).
21
22
23 𝑅𝑚𝑎𝑥 [𝑃]
24 𝑅𝑒𝑞 = [Eq. 10]
25 𝐾𝑑 + [𝑃]
26
27 𝑅𝑚𝑎𝑥1 [𝑃] 𝑅𝑚𝑎𝑥2 [𝑃]
28 𝑅𝑒𝑞 = + [Eq. 11]
29
𝐾𝑑1 + [𝑃] 𝐾𝑑2 + [𝑃]
30
31
32
33 where Req is the response in the equilibrium (maximum experimental measurement for each
34
35 point at different concentrations); [P] is the protein concentration; Rmax is the maximum
36
37
38 surface-binding response for a single site type (calculated from the fitting); Rmax1 and Rmax2 are
39
40 the maximum surface-binding response when considering two site types; Kd, is the macroscopic
41
42 dissociation constant for a single site type, and Kd1 and Kd2 are the dissociation constants when
43
44
45 considering two site types.
46
47 Laser scanning confocal microscopy (LSCM). At different times of growth, a sample
48
49 of the pneumococcal liquid culture was taken and cells were stained with the
50
51
52
LIVE/DEAD BacLight bacterial viability kit (Molecular Probes) to monitor bacterial
53
54 populations. A 3 μL mixture of 1:1 SYTO 9:propidium iodide mix was added to a 1 mL sample
55
56 of the culture medium and kept at room temperature during 15 min in the dark. Confocal images
57
58 were captured using an inverse laser scanning confocal microscope (Leika TCS-SP2-AOBS-
59
60
41
1
2 UV) with a 63× oil-immersion lens. The excitation/emission wavelengths for SYTO 9 and
3
4 propidium iodide were 488/500-550 nm and 543/600-670 nm, respectively.
5
6
Tertiary structure predictions for LZ-C-LytA protein. Three-dimensional
7
8
9 coordinates corresponding to the sequence containing the predicted leucine zipper of
10
11 Pseudomonas putida PhaF phasin (LZ sequence:
12
13 MGLGVPSRNEIKALHQQVDSLTKQIEKLTGASVTPISSRK) were taken from a previous
14
15
16 PhaF model58 and those related to the C-LytA were obtained from the solved X-ray crystal
17
18 structure (PDB code: 1HCX).33 The SwissPDB Viewer 4.1 program78 was used to connect both
19
20 sequences and to generate the basic model of the LZ-C-LytA monomer and of the dimer linked
21
22
23
through the C-LytA moiety (CT2 variant, see Table S2). Next, the monomeric LZ-C-LytA
24
25 PDB file was submitted to the M-ZDOCK server (http://zdock.umassmed.edu/m-zdock/)60 to
26
27 generate the rest of oligomer conformations involving interactions through the leucine zippers
28
29
(LZ2-4 variants, Table S2). In those cases in which the server was not able to render the
30
31
32 corresponding model, the isolated LZ sequence was submitted to the M-ZDOCK and the
33
34 corresponding generated models were used in turn as templates for the alignment of the full-
35
36 length monomer, making use of the “Iterative Fit” utility of SwissPDB Viewer 4.1. The LZ
37
38
39 templates were subsequently removed from the final model.
40
41 The theoretical hydrodynamic properties for each model were estimated by using the
42
43 HyproPro 10 program,61 and using as parameters a disolvent density of 1 g mL-1, viscosity of
44
45
46
0.01 P, temperature of 20 oC and a specific volume 0.722 mL g-1 (taken from that described for
47
48 C-LytA protein).23
49
50 Phagocytosis assays. The animal experiments were approved by the Animal Care
51
52
Committee of the University Hospital of Göttingen, Germany, and by the Niedersächsisches
53
54
55 Landesamt für Verbraucherschutz und Lebensmittelsicherheit (LAVES), Braunschweig,
56
57 Lower Saxony, Germany. The C57/BI6N mice (7 animals for C-Cpl1 and 5 for LzClytA
58
59 experiments, all 10-14 weeks old) were sacrificed by CO2 inhalation. Peritoneal lavage was
60
42
1
2 performed twice with 5 mL of sterile and cool 10 mM phosphate buffered saline (PBS) ( 138
3
4 mM NaCl, 2.7 mM KCl, pH 7.4) using a 21-gauge needle. The peritoneal fluid was then
5
6
collected, centrifuged for 8 min at 500 x g and 4 oC, and the pellet was washed twice with PBS.
7
8
9 The final sediment was suspended in Dulbecco´s modified Eagle´s medium (DMEM) (Gibco,
10
11 Karlsruhe, Germany) suplemented with 10 % of heat-inactivated fetal calf serum (FCS), 100
12
13 U/mL penicillin and 100 µg/mL streptomycin. Cells were counted with a hemocytometer,
14
15
16 plated in a 96-well plate at a density of 15 x 104 cells/well and maintained at 37 °C in a
17
18 humidified atmosphere with 5 % CO2. After 48 h, the culture medium was changed to non-
19
20 antibiotic conditions.
21
22
23
The pneumococcal culture (R6 strain) was grown in tryptic soy broth (Sigma-Aldrich)
24
25 at 37 °C without agitation and in the presence of 5 % CO2. Bacterial growth was monitored by
26
27 quantitative plating of viable bacteria (serial ten-fold dilutions performed in sterile saline
28
29
solution were plated on blood agar, and the bacterial colonies in the appropriate dilutions were
30
31
32 counted). The protein C-Cpl1 (18 µg/mL in 20 mM sodium phosphate, pH 7.0) or LZ-C-LytA
33
34 (20 µg/mL in 20 mM sodium phosphate, pH 7.0) was added to the culture both from the
35
36 beginning of the bacterial growth (co-incubation time, 16 h) or at the late exponential phase
37
38
39 (co-incubation time, 2 h). A control was realized in each case by co-incubation of the culture
40
41 with 20 mM sodium phosphate, pH 7.0.
42
43 A 200 µL sample of the pneumococcal culture previously treated with the respective
44
45
46
protein was centrifuged at room temperature for 5 min at 1,000 × g and the pellet was suspended
47
48 in the same volume of DMEM medium suplemented with 10 % FCS. The bacteria were added
49
50 to the macrophages in a ratio of approximately 100 bacteria per phagocyte (1.5 x 107 CFU/well)
51
52
and co-incubated for 1 h at 37 oC in a humidified atmosphere with 5 % CO2. Next, the
53
54
55 macrophages were washed twice with 0.9 % NaCl, resuspended in DMEM medium
56
57 supplemented with 10 % FCS and 100 mg/mL gentamicin (Sigma-Aldrich) and incubated for
58
59 1 h at 37 oC in the presence of 5 % CO2 in order to kill the ramaining extracellular bacteria.
60
43
1
2 Finally, the macrophages were washed twice with 0.9 % NaCl and lysed by addition of 100 µL
3
4 of distilled water. The bacteria were then quantified by quantitative plating of viable bacteria
5
6
as described above.
7
8
9 The statistical analysis was carried out using GraphPad Prism (GraphPad 6 Software,
10
11 San Diego, CA, USA). Data were expressed as medians (25th percentile/75th percentile).
12
13 Kruskal-Wallis test followed by Dunn’s multiple comparison test was used to compare data.
14
15
16 Probabilities lower than 0.05 were considered statistically significant (****, p < 0.0001).
17
18
19
20
21 ANCILLARY INFORMATION
22
23
24 Corresponding authors: Beatriz Maestro (bmaestro35@gmail.com), Jesús M. Sanz
25 (jmsanz@cib.csic.es)
26
27
28 Present address for Francisco García-Asencio: Developmental Neurobiology Unit, Instituto de
29
30 Neurociencias de Alicante, Miguel Hernández University - Spanish National Research Council
31 (UMH-CSIC), Av. Ramón y Cajal s/n, Sant Joan d'Alacant-03550, Spain
32
33
34
35 Author contributions: Conceived and designed the experiments: ER-M, MS-A, JS, RN, JMS,
36
37
BM. Performed the experiments: ER-M, MS-A, JS, FG-A, BM. Analyzed the data: RN, JMS,
38 BM. Contributed reagents/materials/analysis tools: RN, JMS. Wrote the paper: JS, RN, JMS,
39
40 BM. Contributed with ideas: ER-M, MS-A. J.M.S. and B.M. contributed equally to this work.
41
42
43
Acknowledgments: We thank Cristina Almansa (confocal microscopy), Laura Lagartera (SPR
44
45 experiments), Juan R. Luque (analytical ultracentrifugation) and M. Garzón for skillful
46
47 technical assistance. We especially thank M. Menéndez and P. García for helpful discussions.
48
49 This work was supported by grants from the Spanish Ministry of Economy, Industry and
50 Competitiveness (BIO2013-47684-R and BIO2016-79323-R), the RETICS-FEDER RICET
51
52 RD16/0027/0010 (Spain) and the B. Braun Foundation, Melsungen, Germany. The CIBER de
53
54 Enfermedades Respiratorias (CIBERES) is an initiative of the Spanish Instituto de Salud Carlos
55
56
III.
57
58
59
60
44
1
2 REFERENCES
3
4
5 1. O'Brien, K. L.; Wolfson, L. J.; Watt, J. P.; Henkle, E.; Deloria-Knoll, M.; McCall, N.;
6 Lee, E.; Mulholland, K.; Levine, O. S.; Cherian, T.; Hib and Pneumococcal Global Burden of
7 Disease Study Team, Burden of disease caused by Streptococcus pneumoniae in children
8
younger than 5 years: global estimates. Lancet 2009, 374 (9693), 893-902. DOI:
9
10 10.1016/S0140-6736(09)61204-6.
11
2. Rodrigues, C. M. C.; Groves, H., Community-Acquired Pneumonia in Children: the
12
13
Challenges of Microbiological Diagnosis. J. Clin. Microbiol. 2018, 56 (3). DOI:
14 10.1128/JCM.01318-17.
15
16
3. UNICEF Pneumonia: The Forgotten Killer of Children; United Nations Children’s
17 Emergency Fund: New York, NY, USA, 2006.
18
19
4. Global Burden of Disease Lower Respiratory Infections Collaborators, Estimates of the
20 global, regional, and national morbidity, mortality, and aetiologies of lower respiratory
21 infections in 195 countries, 1990-2016: a systematic analysis for the Global Burden of Disease
22 Study 2016. Lancet Infect. Dis. 2018, 18 (11), 1191-1210. DOI: 10.1016/S1473-
23 3099(18)30310-4.
24
25 5. Tacconelli, E.; Carrara, E.; Savoldi, A.; Harbarth, S.; Mendelson, M.; Monnet, D. L.;
26 Pulcini, C.; Kahlmeter, G.; Kluytmans, J.; Carmeli, Y.; Ouellette, M.; Outterson, K.; Patel, J.;
27 Cavaleri, M.; Cox, E. M.; Houchens, C. R.; Grayson, M. L.; Hansen, P.; Singh, N.;
28 Theuretzbacher, U.; Magrini, N.; W. H. O. Pathogens Priority List Working Group, Discovery,
29
30
research, and development of new antibiotics: the WHO priority list of antibiotic-resistant
31 bacteria and tuberculosis. Lancet Infect. Dis. 2018, 18 (3), 318-327. DOI: 10.1016/S1473-
32 3099(17)30753-3.
33
34 6. Centers for Disease Control and Prevention. Antibiotic resistance threats in the United
35 States; U.S. Department of Health: 2013.
36
37 7. Geno, K. A.; Gilbert, G. L.; Song, J. Y.; Skovsted, I. C.; Klugman, K. P.; Jones, C.;
38 Konradsen, H. B.; Nahm, M. H., Pneumococcal Capsules and Their Types: Past, Present, and
39 Future. Clin. Microbiol. Rev. 2015, 28 (3), 871-99. DOI: 10.1128/CMR.00024-15.
40
41 8. Papadatou, I.; Spoulou, V., Pneumococcal Vaccination in High-Risk Individuals: Are
42 We Doing It Right? Clin. Vaccine Immunol. 2016, 23 (5), 388-395. DOI: 10.1128/CVI.00721-
43 15.
44
45 9. Alexandre, C.; Dubos, F.; Courouble, C.; Pruvost, I.; Varon, E.; Hospital Network for
46 Evaluating the Management of Common Childhood, D.; Martinot, A., Rebound in the
47 incidence of pneumococcal meningitis in northern France: effect of serotype replacement. Acta
48 Paediatr. 2010, 99 (11), 1686-90. DOI: 10.1111/j.1651-2227.2010.01914.x.
49
50 10. Weil-Olivier, C.; van der Linden, M.; de Schutter, I.; Dagan, R.; Mantovani, L.,
51 Prevention of pneumococcal diseases in the post-seven valent vaccine era: a European
52 perspective. BMC Infect. Dis. 2012, 12, 207. DOI: 10.1186/1471-2334-12-207.
53
54 11. Koelman, D. L. H.; Brouwer, M. C.; van de Beek, D., Resurgence of pneumococcal
55 meningitis in Europe and Northern America. Clin. Microbiol. Infect. 2019. DOI:
56
57
10.1016/j.cmi.2019.04.032.
58 12. WHO/UNICEF The WHO/UNICEF estimates of national immunization coverage
59
(WUENIC); 2016.
60
45
1
2 13. Johns Hopkins Bloomberg School of Public Health International Vaccine Access
3 Center (IVAC), VIEW-hub Report: Global Vaccine Introduction and Implementation; 2018.
4
5 14. Todorova, K.; Maurer, P.; Rieger, M.; Becker, T.; Bui, N. K.; Gray, J.; Vollmer, W.;
6 Hakenbeck, R., Transfer of penicillin resistance from Streptococcus oralis to Streptococcus
7 pneumoniae identifies murE as resistance determinant. Mol. Microbiol. 2015, 97 (5), 866-80.
8 DOI: 10.1111/mmi.13070.
9
10 15. Cherazard, R.; Epstein, M.; Doan, T. L.; Salim, T.; Bharti, S.; Smith, M. A.,
11 Antimicrobial Resistant Streptococcus pneumoniae: Prevalence, Mechanisms, and Clinical
12 Implications. Am. J. Ther. 2017, 24 (3), e361-e369. DOI: 10.1097/MJT.0000000000000551.
13
14 16. European Centre for Disease Prevention and Control (ECDC). Surveillance of
15 antimicrobial resistance in Europe; 2017.
16
17 17. Pérez-Dorado, I.; Galán-Bartual, S.; Hermoso, J. A., Pneumococcal surface proteins:
18 when the whole is greater than the sum of its parts. Mol. Oral Microbiol. 2012, 27 (4), 221-45.
19
DOI: 10.1111/j.2041-1014.2012.00655.x.
20
21 18. Maestro, B.; Sanz, J. M., Choline Binding Proteins from Streptococcus pneumoniae: A
22
Dual Role as Enzybiotics and Targets for the Design of New Antimicrobials. Antibiotics
23
24 (Basel) 2016, 5 (2), 21. DOI: 10.3390/antibiotics5020021.
25
19. Giudicelli, S.; Tomasz, A., Attachment of pneumococcal autolysin to wall teichoic
26
27
acids, an essential step in enzymatic wall degradation. J. Bacteriol. 1984, 158 (3), 1188-90.
28 20. Briese, T.; Hakenbeck, R., Interaction of the pneumococcal amidase with lipoteichoic
29
30
acid and choline. Eur. J. Biochem. 1985, 146 (2), 417-27.
31 21. Monterroso, B.; Saiz, J. L.; Garcia, P.; Garcia, J. L.; Menendez, M., Insights into the
32
33
structure-function relationships of pneumococcal cell wall lysozymes, LytC and Cpl-1. J. Biol.
34 Chem. 2008, 283 (42), 28618-28. DOI: 10.1074/jbc.M802808200.
35
22. Medrano, F. J.; Gasset, M.; Lopez-Zumel, C.; Usobiaga, P.; Garcia, J. L.; Menéndez,
36
37 M., Structural characterization of the unligated and choline-bound forms of the major
38 pneumococcal autolysin LytA amidase. Conformational transitions induced by temperature. J.
39 Biol. Chem. 1996, 271 (46), 29152-61.
40
41 23. Usobiaga, P.; Medrano, F. J.; Gasset, M.; García, J. L.; Saiz, J. L.; Rivas, G.; Laynez,
42 J.; Menéndez, M., Structural organization of the major autolysin from Streptococcus
43 pneumoniae. J. Biol. Chem. 1996, 271 (12), 6832-8.
44
45 24. Sanz, J. M.; García, J. L.; Laynez, J.; Usobiaga, P.; Menéndez, M., Thermal stability
46 and cooperative domains of CPL1 lysozyme and its NH2- and COOH-terminal modules.
47 Dependence on choline binding. J. Biol. Chem. 1993, 268 (9), 6125-30.
48
49 25. Maestro, B.; González, A.; García, P.; Sanz, J. M., Inhibition of pneumococcal choline-
50 binding proteins and cell growth by esters of bicyclic amines. FEBS J. 2007, 274 (2), 364-76.
51 DOI: 10.1111/j.1742-4658.2006.05584.x.
52
53 26. Hernández-Rocamora, V. M.; Maestro, B.; de Waal, B.; Morales, M.; García, P.;
54 Meijer, E. W.; Merkx, M.; Sanz, J. M., Multivalent choline dendrimers as potent inhibitors of
55 pneumococcal cell-wall hydrolysis. Angew. Chem. Int. Ed. Engl. 2009, 48 (5), 948-51. DOI:
56
57
10.1002/anie.200803664.
58 27. Ribes, S.; Riegelmann, J.; Redlich, S.; Maestro, B.; de Waal, B.; Meijer, E. W.; Sanz,
59
60
J. M.; Nau, R., Multivalent choline dendrimers increase phagocytosis of Streptococcus
46
1
2 pneumoniae R6 by microglial cells. Chemotherapy 2013, 59 (2), 138-42. DOI:
3 10.1159/000353439.
4
5 28. Nau, R.; Eiffert, H., Minimizing the release of proinflammatory and toxic bacterial
6 products within the host: a promising approach to improve outcome in life-threatening
7 infections. FEMS Immunol. Med. Microbiol. 2005, 44 (1), 1-16. DOI:
8 10.1016/j.femsim.2005.01.001.
9
10 29. Brown, L. A.; Mitchell, A. M.; Mitchell, T. J., Streptococcus pneumoniae and lytic
11 antibiotic therapy: are we adding insult to injury during invasive pneumococcal disease and
12 sepsis? J. Med. Microbiol. 2017. DOI: 10.1099/jmm.0.000545.
13
14 30. Martner, A.; Skovbjerg, S.; Paton, J. C.; Wold, A. E., Streptococcus pneumoniae
15 autolysis prevents phagocytosis and production of phagocyte-activating cytokines. Infect.
16
17
Immun. 2009, 77 (9), 3826-37. DOI: 10.1128/IAI.00290-09.
18 31. Sánchez-Puelles, J. M.; Sanz, J. M.; García, J. L.; García, E., Cloning and expression
19
of gene fragments encoding the choline-binding domain of pneumococcal murein hydrolases.
20
21 Gene 1990, 89 (1), 69-75.
22
32. Dalia, A. B.; Weiser, J. N., Minimization of bacterial size allows for complement
23
24 evasion and is overcome by the agglutinating effect of antibody. Cell Host Microbe 2011, 10
25 (5), 486-96. DOI: 10.1016/j.chom.2011.09.009.
26
27
33. Fernández-Tornero, C.; López, R.; García, E.; Giménez-Gallego, G.; Romero, A., A
28 novel solenoid fold in the cell wall anchoring domain of the pneumococcal virulence factor
29 LytA. Nat. Struct. Biol. 2001, 8 (12), 1020-4. DOI: 10.1038/nsb724.
30
31 34. Maestro, B.; Sanz, J. M., Accumulation of partly folded states in the equilibrium
32 unfolding of the pneumococcal choline-binding module C-LytA. Biochem. J. 2005, 387 (Pt 2),
33 479-88. DOI: 10.1042/BJ20041194.
34
35 35. Guiral, S.; Mitchell, T. J.; Martin, B.; Claverys, J. P., Competence-programmed
36 predation of noncompetent cells in the human pathogen Streptococcus pneumoniae: genetic
37 requirements. Proc. Natl. Acad. Sci. U. S. A. 2005, 102 (24), 8710-5. DOI:
38 10.1073/pnas.0500879102.
39
40 36. García, J. L.; García, E.; Arrarás, A.; García, P.; Ronda, C.; López, R., Cloning,
41 purification, and biochemical characterization of the pneumococcal bacteriophage Cp-1 lysin.
42 J. Virol. 1987, 61 (8), 2573-80.
43
44 37. Hermoso, J. A.; Monterroso, B.; Albert, A.; Galán, B.; Ahrazem, O.; García, P.;
45 Martínez-Ripoll, M.; García, J. L.; Menéndez, M., Structural basis for selective recognition of
46
pneumococcal cell wall by modular endolysin from phage Cp-1. Structure 2003, 11 (10), 1239-
47
48 49.
49
38. Pérez-Dorado, I.; González, A.; Morales, M.; Sanles, R.; Striker, W.; Vollmer, W.;
50
51 Mobashery, S.; García, J. L.; Martínez-Ripoll, M.; García, P.; Hermoso, J. A., Insights into
52 pneumococcal fratricide from the crystal structures of the modular killing factor LytC. Nat.
53 Struct. Mol. Biol. 2010, 17 (5), 576-81. DOI: 10.1038/nsmb.1817.
54
55 39. Li, Q.; Cheng, W.; Morlot, C.; Bai, X. H.; Jiang, Y. L.; Wang, W.; Roper, D. I.; Vernet,
56 T.; Dong, Y. H.; Chen, Y.; Zhou, C. Z., Full-length structure of the major autolysin LytA. Acta
57 Crystallogr. D Biol. Crystallogr. 2015, 71 (Pt 6), 1373-81. DOI:
58 10.1107/S1399004715007403.
59
60
47
1
2 40. García, E.; García, J. L.; García, P.; Arrarás, A.; Sánchez-Puelles, J. M.; López, R.,
3 Molecular evolution of lytic enzymes of Streptococcus pneumoniae and its bacteriophages.
4 Proc. Natl. Acad. Sci. U. S. A. 1988, 85 (3), 914-8.
5
6 41. Buey, R. M.; Monterroso, B.; Menéndez, M.; Diakun, G.; Chacón, P.; Hermoso, J. A.;
7 Díaz, J. F., Insights into molecular plasticity of choline binding proteins (pneumococcal surface
8 proteins) by SAXS. J. Mol. Biol. 2007, 365 (2), 411-24. DOI: 10.1016/j.jmb.2006.09.091.
9
10 42. Gosink, K. K.; Mann, E. R.; Guglielmo, C.; Tuomanen, E. I.; Masure, H. R., Role of
11 novel choline binding proteins in virulence of Streptococcus pneumoniae. Infect. Immun. 2000,
12 68 (10), 5690-5.
13
14 43. Waterhouse, A.; Bertoni, M.; Bienert, S.; Studer, G.; Tauriello, G.; Gumienny, R.; Heer,
15 F. T.; de Beer, T. A. P.; Rempfer, C.; Bordoli, L.; Lepore, R.; Schwede, T., SWISS-MODEL:
16
17
homology modelling of protein structures and complexes. Nucleic Acids Res. 2018, 46 (W1),
18 W296-W303. DOI: 10.1093/nar/gky427.
19
44. Xu, Q.; Zhang, J. W.; Chen, Y.; Li, Q.; Jiang, Y. L., Crystal structure of the choline-
20
21 binding protein CbpJ from Streptococcus pneumoniae. Biochem. Biophys. Res. Commun. 2019,
22 514 (4), 1192-1197. DOI: 10.1016/j.bbrc.2019.05.053.
23
24 45. Grosdidier, A.; Zoete, V.; Michielin, O., SwissDock, a protein-small molecule docking
25 web service based on EADock DSS. Nucleic Acids Res. 2011, 39 (Web Server issue), W270-
26 7. DOI: 10.1093/nar/gkr366.
27
28 46. Greenfield, N. J., Using circular dichroism collected as a function of temperature to
29 determine the thermodynamics of protein unfolding and binding interactions. Nat. Protoc.
30 2006, 1 (6), 2527-35. DOI: 10.1038/nprot.2006.204.
31
32 47. Fukada, H.; Sturtevant, J. M.; Quiocho, F. A., Thermodynamics of the binding of L-
33 arabinose and of D-galactose to the L-arabinose-binding protein of Escherichia coli. J. Biol.
34 Chem. 1983, 258 (21), 13193-8.
35
36 48. Freire, E.; van Osdol, W. W.; Mayorga, O. L.; Sánchez-Ruiz, J. M., Calorimetrically
37 determined dynamics of complex unfolding transitions in proteins. Annu. Rev. Biophys.
38 Biophys. Chem. 1990, 19, 159-88. DOI: 10.1146/annurev.bb.19.060190.001111.
39
40 49. Pace, C. N.; Scholtz, J. M., Measuring the conformational stability of a protein. In
41 Protein Structure: A Practical Approach, Creighton, T. E., Ed. Oxford University Press: 1989;
42 pp 299-321.
43
44 50. Myers, J. K.; Pace, C. N.; Scholtz, J. M., Denaturant m values and heat capacity
45 changes: relation to changes in accessible surface areas of protein unfolding. Protein Sci. 1995,
46
4 (10), 2138-48. DOI: 10.1002/pro.5560041020.
47
48 51. Spolar, R. S.; Livingstone, J. R.; Record, M. T., Jr., Use of liquid hydrocarbon and
49
amide transfer data to estimate contributions to thermodynamic functions of protein folding
50
51 from the removal of nonpolar and polar surface from water. Biochemistry 1992, 31 (16), 3947-
52 55. DOI: 10.1021/bi00131a009.
53
54
52. Sanz, J. M.; López, R.; García, J. L., Structural requirements of choline derivatives for
55 'conversion' of pneumococcal amidase. A new single-step procedure for purification of this
56 autolysin. FEBS Lett. 1988, 232 (2), 308-12.
57
58 53. Sánchez-Puelles, J. M.; Sanz, J. M.; García, J. L.; García, E., Immobilization and single-
59 step purification of fusion proteins using DEAE-cellulose. Eur. J. Biochem. 1992, 203 (1-2),
60 153-9.
48
1
2 54. Silva-Martin, N.; Retamosa, M. G.; Maestro, B.; Bartual, S. G.; Rodes, M. J.; García,
3 P.; Sanz, J. M.; Hermoso, J. A., Crystal structures of CbpF complexed with atropine and
4 ipratropium reveal clues for the design of novel antimicrobials against Streptococcus
5 pneumoniae. Biochim. Biophys. Acta 2014, 1840 (1), 129-35. DOI:
6
10.1016/j.bbagen.2013.09.006.
7
8 55. Maestro, B.; Velasco, I.; Castillejo, I.; Arévalo-Rodriguez, M.; Cebolla, A.; Sanz, J. M.,
9
Affinity partitioning of proteins tagged with choline-binding modules in aqueous two-phase
10
11 systems. J. Chromatogr. A 2008, 1208 (1-2), 189-96. DOI: 10.1016/j.chroma.2008.08.106.
12 56. De Las Rivas, B.; García, J. L.; López, R.; García, P., Purification and polar localization
13
14
of pneumococcal LytB, a putative endo-beta-N-acetylglucosaminidase: the chain-dispersing
15 murein hydrolase. J. Bacteriol. 2002, 184 (18), 4988-5000.
16
17
57. Retamosa, M. G.; Díez-Martínez, R.; Maestro, B.; García-Fernández, E.; de Waal, B.;
18 Meijer, E. W.; García, P.; Sanz, J. M., Aromatic Esters of Bicyclic Amines as Antimicrobials
19 against Streptococcus pneumoniae. Angew. Chem. Int. Ed. Engl. 2015, 54 (46), 13673-7. DOI:
20 10.1002/anie.201505700.
21
22 58. Maestro, B.; Galán, B.; Alfonso, C.; Rivas, G.; Prieto, M. A.; Sanz, J. M., A new family
23 of intrinsically disordered proteins: structural characterization of the major phasin PhaF from
24 Pseudomonas putida KT2440. PLoS One 2013, 8 (2), e56904. DOI:
25 10.1371/journal.pone.0056904.
26
27 59. Tarazona, N. A.; Maestro, B.; Revelles, O.; Sanz, J. M.; Prieto, M. A., Role of leucine
28 zipper-like motifs in the oligomerization of Pseudomonas putida phasins. Biochim Biophys
29 Acta Gen Subj 2019, 1863 (2), 362-370. DOI: 10.1016/j.bbagen.2018.11.002.
30
31 60. Pierce, B. G.; Wiehe, K.; Hwang, H.; Kim, B. H.; Vreven, T.; Weng, Z., ZDOCK
32 server: interactive docking prediction of protein-protein complexes and symmetric multimers.
33
Bioinformatics 2014, 30 (12), 1771-3. DOI: 10.1093/bioinformatics/btu097.
34
35 61. Ortega, A.; Amorós, D.; García de la Torre, J., Prediction of hydrodynamic and other
36
solution properties of rigid proteins from atomic- and residue-level models. Biophys. J. 2011,
37
38 101 (4), 892-8. DOI: 10.1016/j.bpj.2011.06.046.
39 62. Varea, J.; Saiz, J. L.; Lopez-Zumel, C.; Monterroso, B.; Medrano, F. J.; Arrondo, J. L.;
40
41
Iloro, I.; Laynez, J.; García, J. L.; Menéndez, M., Do sequence repeats play an equivalent role
42 in the choline-binding module of pneumococcal LytA amidase? J. Biol. Chem. 2000, 275 (35),
43 26842-55. DOI: 10.1074/jbc.M004379200.
44
45 63. García, J. L.; Díaz, E.; Romero, A.; García, P., Carboxy-terminal deletion analysis of
46 the major pneumococcal autolysin. J. Bacteriol. 1994, 176 (13), 4066-72.
47
48 64. López, R.; García, E.; García, P.; García, J. L., The pneumococcal cell wall degrading
49 enzymes: a modular design to create new lysins? Microb. Drug Resist. 1997, 3 (2), 199-211.
50
51 65. Yoon, H. R.; Choi, H.; Choi, Y. A.; Kim, J. A.; Jung, J.; Kim, H. M.; Jung, Y.,
52 Fabrication of Oligomeric Avidin Scaffolds for Valency-Controlled Surface Display of
53 Functional Ligands. Angew. Chem. Int. Ed. Engl. 2018, 57 (38), 12410-12414. DOI:
54 10.1002/anie.201805749.
55
56 66. Choi, S.-K., Synthetic Multivalent Molecules: Concepts and Biomedical Applications.
57 Wiley-Interscience: New York (USA), 2004.
58
59 67. Fasting, C.; Schalley, C. A.; Weber, M.; Seitz, O.; Hecht, S.; Koksch, B.; Dernedde, J.;
60 Graf, C.; Knapp, E. W.; Haag, R., Multivalency as a chemical organization and action principle.
Angew. Chem. Int. Ed. Engl. 2012, 51 (42), 10472-98. DOI: 10.1002/anie.201201114.
49
1
2 68. Margni, R. A.; Parma, E. A.; Cerone, S.; Erpelding, A.; Perdigon, G., Agglutinating
3 and non-agglutinating antibodies in rabbits inoculated with a particulate antigen (Salmonella
4 typhimurium). Immunology 1983, 48 (2), 351-9.
5
6 69. Bull, C. G., The Agglutination of Bacteria in vivo. J. Exp. Med. 1915, 22, 484-491.
7
8 70. Brent, R.; Ptashne, M., Mechanism of action of the lexA gene product. Proc. Natl. Acad.
9 Sci. U. S. A. 1981, 78 (7), 4204-8. DOI: 10.1073/pnas.78.7.4204.
10
11 71. Sambrook, J.; Russell, D. W., Molecular cloning: a laboratory manual. Cold Spring
12 Harbor Laboratory Press: Cold Spring Harbor, N.Y., 2001.
13
14 72. Hoskins, J.; Alborn, W. E., Jr.; Arnold, J.; Blaszczak, L. C.; Burgett, S.; DeHoff, B. S.;
15 Estrem, S. T.; Fritz, L.; Fu, D. J.; Fuller, W.; Geringer, C.; Gilmour, R.; Glass, J. S.; Khoja, H.;
16 Kraft, A. R.; Lagace, R. E.; LeBlanc, D. J.; Lee, L. N.; Lefkowitz, E. J.; Lu, J.; Matsushima,
17 P.; McAhren, S. M.; McHenney, M.; McLeaster, K.; Mundy, C. W.; Nicas, T. I.; Norris, F. H.;
18 O'Gara, M.; Peery, R. B.; Robertson, G. T.; Rockey, P.; Sun, P. M.; Winkler, M. E.; Yang, Y.;
19
Young-Bellido, M.; Zhao, G.; Zook, C. A.; Baltz, R. H.; Jaskunas, S. R.; Rosteck, P. R., Jr.;
20
21 Skatrud, P. L.; Glass, J. I., Genome of the bacterium Streptococcus pneumoniae strain R6. J.
22 Bacteriol. 2001, 183 (19), 5709-17. DOI: 10.1128/JB.183.19.5709-5717.2001.
23
24 73. Lacks, S.; Hotchkiss, R. D., A study of the genetic material determining an enzyme in
25 Pneumococcus. Biochim. Biophys. Acta 1960, 39, 508-18.
26
27
74. Pettersen, E. F.; Goddard, T. D.; Huang, C. C.; Couch, G. S.; Greenblatt, D. M.; Meng,
28 E. C.; Ferrin, T. E., UCSF Chimera - a visualization system for exploratory research and
29 analysis. J. Comput. Chem. 2004, 25 (13), 1605-12. DOI: 10.1002/jcc.20084.
30
31 75. Schuck, P., Size-distribution analysis of macromolecules by sedimentation velocity
32 ultracentrifugation and Lamm equation modeling. Biophys. J. 2000, 78 (3), 1606-19. DOI:
33 10.1016/S0006-3495(00)76713-0.
34
35 76. van Holde, K. E., Sedimentation in Physical Biochemistry. Prentice Hall, Inc.:
36 Englewood Cliffs, NJ, USA, 1986.
37
38 77. Laue, T. M.; Shah, B. D.; Ridgeway, T. M.; Pelletier, S. L., Computer-aided
39 interpretation of analytical sedimentation data for proteins. In Analytical Ultracentrifugation
40 in Biochemistry and Polymer Science, Harding, S.; Rowe, A.; Horton, J., Eds. Royal Society
41 of Chemistry: Cambridge, UK, 1992; pp 90-125.
42
43 78. Guex, N.; Peitsch, M. C., SWISS-MODEL and the Swiss-PdbViewer: an environment
44 for comparative protein modeling. Electrophoresis 1997, 18 (15), 2714-23. DOI:
45 10.1002/elps.1150181505.
46
47 79. Fraczkiewicz, R.; Braun, W., Exact and Efficient Analytical Calculation of the
48 Accessible Surface Areas and Their Gradients for Macromolecules. J Comp Chem 1998, 19,
49
319-333.
50
51 80. Langmuir, I., The constitution and fundamental properties of solids and liquids. J. Am.
52
Chem. Soc. 1916, 38, 2221-2295.
53
54
55
56
57
58
59
60
50
1
2
3
4
5 SUPPORTING INFORMATION
6
7 Figure S1. Intrinsic fluorescence emission spectra of C-Cpl1 (A) and C-CbpD (B) in phosphate
8
9 buffer (black) plus 20 mM NaCl (green) or 20 mM choline chloride (red); Spectra registered
10 at various choline concentrations for C-Cpl1 (C) and C-CbpD (D).
11
12 Figure S2. A planktonic culture in the stationary phase of S. pneumoniae R6CIB17 in the
13 absence (left) and the presence (right) of C-LytA 1 µM
14
15
16
Figure S3. Upper panel, amino acid sequence and structural model of the monomer of the LZ-
17 C-LytA protein. Sequence derived from the PhaF phasin is shown in blue, with the predicted
18 leucine-zipper motif underlined. Lower panels: three-dimensional models of LZ-C-LytA
19
20
oligomers. CT species, dimerization through the C-terminal hairpin of C-LytA; LZ species,
21 oligomerization through the leucine zipper sequence.
22
23 Table S1. Theoretical estimation of Cp values for C-Cpl1 and C-CbpD proteins. ASA values
24 were calculated from PDB files with the GetArea utilities.79
25
26
27
Table S2. Theoretical prediction of the hydrodynamic properties of modelled LZ-C-LytA
28 oligomers.
29
30 Table S3. Conditions of SPR experiments. Vf, flow rate; tas, association time; tdis, dissociation
31 time.
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
51
1
2 For Table of Contents Use Only
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
52

Artículo Micro Médica

Uploaded by

Copyright:

Available Formats

You might also like

Artículo Micro Médica

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Artículo Micro Médica

Uploaded by

Copyright:

Available Formats

Subscriber access provided by Universiteit Utrecht

is published by the American Chemical Society. 1155 Sixteenth Street N.W.,

37 division binding hydrolisis

You might also like