bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
1
SapM mutation to improve the BCG vaccine: genomic, transcriptomic and preclinical safety
2
characterization
3
Nele Festjensa,b#, Kristof Vandewallea,b,1 , Erica Houthuysa,b, 2, Evelyn Pletsa,b, Dieter Vanderschaeghea,b,3,
4
Katlyn Borgersa,b, Annelies Van Heckea,b, Petra Tielsa,b, Nico Callewaerta,b#
5
a
6
b
7
Belgium
VIB-UGent Center for Medical Biotechnology; Technologiepark 927, 9052 Zwijnaarde – Belgium.
Department of Biochemistry and Microbiology, Ghent University; Technologiepark 927, 9052 Zwijnaarde -
8
9
# Joint corresponding authors
10
Technologiepark 927
11
9052 Zwijnaarde
12
Nele.Festjens@vib-ugent.be
13
Nico.Callewaert@vib-ugent.be
14
Phone: +32 (0) 9 33 13 630
Current address: 1Inbiose, Technologiepark 3, bus 41, 9052 Zwijnaarde; 2iTeos Therapeutics, Rue des Frères
Wright 29, 6041 Gosselies; 3NLO, AA Tower, Technologiepark 19, 9052 Zwijnaarde
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
15
Abstract
16
The Mycobacterium bovis Bacille Calmette Guérin (BCG) vaccine shows variable efficacy in protection against
17
adult tuberculosis (TB). Earlier, we have described a BCG mutant vaccine with a transposon insertion in the
18
gene coding for the secreted acid phosphatase SapM, which led to enhanced long-term survival of vaccinated
19
mice challenged with TB infection. To facilitate development of this mutation as part of a future improved
20
live attenuated TB vaccine, we have now characterized the genome and transcriptome of this sapM::Tn
21
mutant versus parental BCG Pasteur. Furthermore, we show that the sapM::Tn mutant had an equal low
22
pathogenicity as WT BCG upon intravenous administration to immunocompromised SCID mice, passing this
23
important safety test. Subsequently, we investigated the clearance of this improved vaccine strain following
24
vaccination and found a more effective innate immune control over the sapM::Tn vaccine bacteria as
25
compared to WT BCG. This leads to a fast contraction of IFNγ producing Th1 and Tc1 cells after sapM::Tn BCG
26
vaccination. These findings corroborate that a live attenuated vaccine that affords improved long-term
27
survival upon TB infection can be obtained by a mutation that further attenuates BCG. These findings suggest
28
that an analysis of the effectiveness of innate immune control of the vaccine bacteria could be instructive
29
also for other live attenuated TB vaccines that are currently under development, and encourage further
30
studies of SapM mutation as a strategy in developing a more protective live attenuated TB vaccine.
2
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
31
Keywords
32
Tuberculosis, vaccine, SapM, immunomodulation
3
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
33
1. Introduction
34
Today, TB remains one of the major causes of infectious disease and death throughout the world [1]. About
35
10 million people developed TB in 2017, with an estimated 1.3 million deaths.
36
Although new and promising anti-tubercular drugs are in or approaching the clinic, full control over this
37
devastating disease will necessarily require a good, preventive anti-TB vaccine [2]. M. bovis Bacille-Calmette-
38
Guérin (BCG) is the only licensed vaccine today, and already in use for almost 100 years. Although BCG is
39
protective against TB meningitis and disseminated TB in children, its efficacy against pulmonary TB is highly
40
variable and not sufficient for disease control [3–5].
41
Currently, 13 vaccine candidates are under active development and in the process of clinical testing. Besides
42
two live recombinant candidates (VPM1002 and MTBVAC) which are being tested to replace BCG as a priming
43
vaccine, the candidates being mainly variations of booster vaccines with recurrent key mycobacterial
44
antigens [2,6,7]. Several of these vaccine candidates yield enhanced interferon γ (IFNɣ)-producing cellular
45
immunity and enhanced control over bacterial replication in the first weeks after TB challenge in animal
46
models. However, reports that show extended long-term survival of vaccinated animals upon TB challenge
47
are much rarer. In the category of priming vaccines, it has been shown that rBCG30 (BCG Tice overexpressing
48
the M. tb 30-kDa major secretory protein antigen 85B)-immunized mice survive longer than BCG-immunized
49
mice following TB challenge [8]. SO2 (phoP inactivated in M. tb strain MT103), the MTBVAC (containing a
50
double deletion in phoP and fadD26) prototype, conferred greater efficacy than BCG in a high-dose challenge,
51
long-term survival experiment in guinea pigs, while no difference to WT BCG was observed after low-dose
52
challenge [9]. The recombinant BCG strain AFRO-1 (BCG1331 ΔureC::ΩpfoAG137Q, Ag85A, Ag85B and TB10.4)
53
affords better protection than BCG in mice following aerosol challenge with Mycobacterium tuberculosis
54
(M.tb) [10]. Additionally, some prime-boost vaccination strategies demonstrated improved long-term
55
survival of M.tb-challenged guinea pigs versus BCG alone: BCG priming followed by the M72 recombinant
56
protein booster vaccine (fusion of Mtb32 and Mtb39 antigens of Mtb) [11] and ID93 (combines four antigens
4
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
57
belonging to families of M.tb proteins associated with virulence (Rv2608, Rv3619, Rv3620) or latency
58
(Rv1813)/GLA-SE)) [12].
59
Earlier, we have demonstrated how a sapM transposon mutant of M. bovis BCG, hereafter called sapM::Tn
60
BCG, resulted in a TB vaccine candidate that provides improved long-term survival in mice upon both systemic
61
and intratracheal challenge with M. tb as compared to the parental BCG Pasteur-derived strain (hereafter
62
called WT BCG) [13]. We suggested that the secreted acid phosphatase SapM may have evolved as an
63
important immunomodulatory protein of Mycobacteria. The importance of SapM for mycobacterial
64
immunomodulation was since confirmed by independent laboratories [14–17]. In the present study, we
65
report on the in-depth genomic and transcriptomic analysis of the sapM::Tn BCG mutant, to assess which
66
alterations are induced by the transposon insertion in the sapM locus.
67
One of the major impediments to TB vaccine development is the incomplete understanding of the
68
mechanisms of protective immunity against M.tb, so far obstructing rational vaccine development [18].
69
Research on animal models and ongoing clinical trials may reveal markers that correlate with a vaccine’s
70
protective potential. However, as long as we lack these correlates, vaccine design will be challenging. One
71
general idea that dominated immunology of infectious diseases over the last decades, is the T helper type
72
1/type2 (Th1/Th2) paradigm. Following this model, Th1 cells would protect the host from intracellular
73
pathogens, like M.tb. Indeed, the early appearance of Th1 type CD4+ T cells secreting IFNγ is necessary for
74
the orchestration of protective immunity in the infected lung [19]. However, controversy has arisen about
75
whether the quantitative extent of IFNɣ production in recall experiments has value as a correlate of vaccine-
76
induced protection against TB. Indeed, several reports show that other anti-tuberculosis CD4 T-cell effector
77
functions are at play [20–24].
78
In this evolving context, we also report on the safety in immunocompromised mice as well as rapid innate
79
control over the sapM::Tn vaccine bacteria, with intriguing impact on the induced adaptive immunity. Our
80
findings suggest that a live attenuated TB vaccine that behaves more like an acute, rapidly immune-controlled
5
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
81
infection, rather than the protracted chronic infection caused by the current BCG vaccine, may yield an
82
immune status that affords more prolonged control of a subsequent TB infection.
6
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
83
2. Results
84
2.1 In depth characterization of the sapM::Tn BCG versus WT BCG strain
85
2.1.1
86
The M. bovis BCG Pasteur strain 1721, used to create the sapM::Tn BCG mutant strain [13,25], differs from
87
the 1173P2 Pasteur vaccine strain by a K43R point mutation in the rpsL gene, conferring streptomycine
88
resistance [26], which is useful to avoid contaminations during the lengthy and involved procedures in
89
ordered transposon mutant library production. To investigate whether there are any additional variants
90
between the WT BCG Pasteur and the sapM::Tn BCG mutant, we performed a whole-genome resequencing
91
analysis on both strains (paired end 2 x 150 bp), mapping to the M. bovis BCG Pasteur 1173P reference
92
genome, resulting in ± 80x average coverage of mapped reads (Suppl. Fig. 1). We could reconfirm the location
93
of the inserted Himar1 transposon, at the TA dinucleotide 8 bp before the sapM start codon (Fig. 1). This was
94
also demonstrated by de novo assembly of the sequencing reads from the sapM::Tn BCG mutant (data not
95
shown). In addition, we performed a probabilistic variant analysis on both WT BCG and sapM::Tn BCG
96
mappings and detected only minor modifications (single amino acid change) with unknown impact. These
97
variants are listed in Table 1 and were verified by Sanger sequencing. All variants versus the Pasteur reference
98
are present in both the parental as well as the sapM::Tn strains, with only one exception, i.e. the frameshift
99
mutation in the sugI gene, coding for sugar-transport integral membrane protein. This mutation probably
100
arose after the library preparation (in a later passage of the strain), since this frameshift mutation is not
101
present in sapM::Tn. The exact function of sugI is unknown, however, it shows distant sequence similarity to
102
glucose permease GlcP of S. coelicolor and the galactose (GalP) and arabinose (AraE) transporters of E. coli.
103
Thus, the system is likely to transport a monosaccharide [27]. It is described to be non-essential in the H37Rv
104
strain [28], which explains the absence of phenotype in WT BCG due to the frameshift mutation. The SNV in
105
sapM::Tn BCG SugI also likely arose after library preparation since it is not present in WT BCG. Hence, sugI
106
appears to be a highly variable gene in the BCG lineage during in vitro cultivation, as the three strains each
107
have a different sequence variant. Interestingly, both the parental as well as the sapM::Tn strain contain a
Whole-genome resequencing
7
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
108
frame-shift mutation in the gene coding for FadD26, a key enzyme in the biosynthesis pathway of the
109
phthiocerol dimycocerosate (PDIM) class of virulence lipids. This fadD26 mutation is interesting from a
110
vaccine engineering point-of-view, as it further attenuates virulence [29,30].
111
2.1.2
112
Transcriptome analysis, by both RNAseq analysis and RT-PCR, demonstrated that the sapM transcript is ~20-
113
50-fold reduced compared to the WT BCG, while the transcription levels of other genes remain largely
114
unchanged (< 2-fold change), except for a ~10-fold upregulation of upp (encoding uracil
115
phosphoribosyltransferase) which is immediately upstream of sapM in the genome, and a modest 2.4-fold
116
downregulation of its downstream gene (BCG_3376) (Fig. 2A-C, Suppl. Fig. 2A-B, Suppl. Table 1). A few (~6)
117
other genes show a differential expression pattern (~2-fold change) between WT BCG and sapM::Tn,
118
however, this effect is marginal compared to the differences in expression of sapM and upp. The transposon
119
insertion thus only majorly affects one nearby gene, and the changes in sapM and upp transcription have
120
almost no secondary effects on overall transcriptional regulation. According to the Operon Correlation
121
Browser on the TB Database [31], the gene coding for SapM in M. tb (Rv3310) and the downstream gene
122
Rv3311 potentially form an operon. However, in M. bovis BCG we see that transposon disruption of the sapM
123
upstream region in sapM::Tn BCG very strongly reduces the levels of sapM transcript, while having a much
124
more modest effect on the levels of BCG_3376 (Rv3311 ortholog) transcription. These data indicate that both
125
genes are likely differentially regulated, questioning the idea of an operonic structure.
126
2.1.3
127
As the SapM protein contains a 43 AA N-terminal signal sequence and thus gets secreted in the culture
128
medium, we collected culture supernatant samples at various time-points.
129
First, we performed an ELISA on these supernatant samples with a newly generated antibody directed against
130
the recombinant SapM protein produced in E. coli (Fig. 2D). While the amount of SapM protein steadily
131
increases over time of culture in the WT BCG, the signal is nearly absent in the sapM::Tn BCG mutant samples
Transposon-induced transcriptional effects
SapM protein is undetectable and secreted total phosphatase activity is strongly reduced
8
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
132
for all time-points. We further confirmed the absence of full-length SapM protein in the sapM::Tn BCG strain
133
by SDS-PAGE and western blotting in an independent experiment (Fig. 2E). We detected a band of ± 28 kDa
134
in the WT BCG, but not in the sapM::Tn BCG mutant sample. A smear is also detected above the 28 kDa band,
135
which likely represents aspecific binding (as the pattern appears to be identical for all samples). To check for
136
any remaining phosphatase activity in the culture medium, we performed an in vitro phosphatase assay with
137
p-Nitrophenyl phosphate (pNPP) as a substrate [32] (Fig. 2F). The measured in vitro activity is significantly
138
reduced for the sapM::Tn BCG mutant sample compared to WT BCG at early time-points. At later time-points
139
(day 8), we still observe a considerable signal, which is expected, as the assay would also pick up activity of
140
other phosphatases such as the protein tyrosine phosphatases (PtpA and PtpB) [33].
141
Consequently, based on all of the data on this vaccine strain, we concluded at this point that its improved
142
vaccine efficiency must indeed be due to a loss of function of SapM or the upregulation of upp transcription,
143
or both.
144
2.2 In vivo safety of sapM::Tn BCG versus WT BCG
145
A requirement of live attenuated vaccines is that they should be safe, even in immunocompromised hosts.
146
Our previous analysis of bacterial replication in immunocompetent Balb/c mice, infected intravenously, did
147
not show any difference between WT BCG and sapM::Tn BCG [13]. The number of granulomas formed in
148
livers and lungs post-infection was unaffected [13]. To unambiguously demonstrate safety of the sapM::Tn
149
BCG mutant compared to WT BCG in vivo, we investigated bacterial virulence in the absence of adaptive
150
immunity. Immunocompromised SCID mice, lacking both T and B cells, were infected intravenously (i.v.) with
151
a high (3x107 CFU) or lower dose (3x106 CFU) of both strains and survival was monitored (Fig. 3). After both
152
low and high doses, SCID mice infected with sapM::Tn BCG mutant showed similar survival to mice infected
153
with WT BCG. Thus, in the SCID model, virulence of sapM::Tn BCG is comparable to WT BCG.
154
2.3 At the lymph node draining the vaccination site, the sapM::Tn BCG mutant is controlled faster than
155
the WT BCG strain
9
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
156
Our previously reported observations indicated that vaccination with the sapM::Tn BCG mutant triggered
157
more effective recruitment of iDCs into the draining lymph nodes (LNs) in Balb/c mice [13]. This lead us to
158
study the influence of this more effective induction of the innate immune response at early time points after
159
sapM::Tn BCG vaccination on the fate of the vaccine inoculum. To enhance the validity, experiments were
160
performed in F1 mice (C57BL/6J x Balb/c) to assess robustness of the observed effects to a more
161
heterogeneous immunogenetic background [34]. After subcutaneous (s.c) vaccination of F1 mice, less
162
bacteria (~50%) were recovered from LNs of sapM::Tn BCG infected mice as compared to WT BCG-infected
163
mice at all time-points analyzed (i.e. days 8, 14 and 28 post-infection) (Fig. 4A). It is clear that whereas WT
164
BCG could amplify from the inoculum, this was barely the case for the sapM::Tn BCG mutant. Since
165
differences are clear already 8 days post-vaccination, this result is most consistent with improved control by
166
innate immune mechanisms. Indeed, an analysis of intracellular survival of WT BCG and sapM::Tn BCG upon
167
infection of bone marrow derived macrophages (BM-DMs), showed that WT BCG and sapM::Tn BCG are
168
taken up similarly (as shown before [13]) but that growth of the sapM::Tn BCG mutant is better controlled
169
compared to the WT BCG strain upon infection (Fig. 4B). In line with these findings, such improved clearance
170
of SapM-mutated M.tb by BM-DMs has recently also been described [16], an effect that was attributed to
171
improved phagosomal maturation.
172
To further evaluate the impact of early innate growth control of the bacteria, we have analyzed survival of
173
sapM::Tn BCG and WT BCG after intravenous infection of F1 mice. Twenty-four hours post-infection, the
174
bacterial load in the lungs and spleens from mice that were infected with WT BCG was higher than the loads
175
of those infected with the sapM::Tn BCG mutant (~50%) (Fig. 4C). Three weeks post-infection, the difference
176
was still observed in the lungs, but not in the spleens. Similar results were obtained in the spleen of C57BL/6J
177
mice, but not in the lungs (Suppl. Fig. 3A). Cytokine levels were analyzed in the serum of i.v. exposed mice at
178
different time-points (6h, 24h, 48h post-vaccination), and significantly higher levels of IL1-β and IL17 could
179
be measured 24h post-sapM::Tn BCG vaccination compared to WT BCG vaccination (Suppl. Fig. 3B). In view
180
of this early time point post-vaccination, both IL1-β and IL17 are produced by innate immune cells. These
10
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
181
particular cytokines are known to be important innate cytokines for bacterial killing, particularly in
182
mycobacterial infection [35,36], which is consistent with the reduced sapM::Tn BCG bacterial load.
183
2.4 Complementation of the sapM::Tn mutation reverses the phenotypes.
184
As the Tn insertion in the sapM locus caused both a loss of function of sapM and a strong upregulation of the
185
upstream upp gene, we assessed whether the improved early innate growth control of the sapM::Tn BCG
186
bacteria was due to the loss of function of SapM or the upregulation of upp transcription. For this purpose,
187
we have complemented the sapM::Tn BCG mutant strain with the sapM gene under control of its own
188
promotor. The sapM expression is reverted to wild type levels in this sapM::Tn:compl BCG mutant, but upp
189
levels are still increased similarly as in the sapM::Tn BCG mutant (Fig. 5). Analysis of bacterial load in the
190
lungs and spleens from mice that were infected with either sapM::Tn:compl BCG mutant, sapM::Tn BCG or
191
WT BCG showed that complementation of sapM expression abolished the improved growth control of the
192
sapM::Tn BCG mutant compared to the WT BCG strain upon infection (Fig. 6). These results confirm that the
193
observed phenotypes are due to the reduced expression of sapM and not because of upregulation of upp
194
(Fig. 2C).
195
2.5 Faster kinetics of iDC recruitment to lymphoid organs and decrease in IFNγ-producing CD4 + and CD8+ T
196
recall response in sapM::Tn BCG vaccinated mice compared to WT BCG vaccinated mice
197
We then further analyzed how improved innate control over vaccine bacteria changes the immune response
198
to the vaccine. Hereto, we have vaccinated C57BL/6J mice s.c. with WT BCG or sapM::Tn BCG mutant. At
199
different time points, mice were sacrificed and LNs and spleens were isolated. DC phenotyping was
200
performed. Significant differences were observed in total leukocyte numbers and iDC recruitment kinetics
201
after sapM::Tn BCG versus WT BCG vaccination. In general, we observe a faster induction of the innate
202
immune response, reflected by a faster recruitment of innate inflammatory cells to the secondary lymphoid
203
organs (dLN-spleen), and by a faster contraction of the innate immune response, seen by lower leukocyte
204
cell numbers at day 14 post-vaccination with BCG sapM::Tn BCG (Fig. 7). Next, T cell immunity was assessed
205
by restimulation of splenocytes or LN cells with either the mycobacterial antigen Ag85A, the M.tb extract
11
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
206
PPD or by anti-CD3/28, a stimulus that activates all T cells, as a control. Vaccination with sapM::Tn BCG leads
207
to a lower number of lymph node and spleen IFNγ-producing CD4+ (Th1) and CD8+ (Tc1) T cells 2 weeks post-
208
vaccination compared to WT BCG (Fig. 8A). After vaccination with a lower dose, this observation is only seen
209
for Tc1 cells, following polyclonal stimulation in the LNs and spleen (Suppl. Fig. 4). IFNγ measurements in the
210
supernatant following stimulation with the different antigens also demonstrated lower levels of IFNγ in the
211
sapM::Tn BCG condition (Fig. 8B). These lower numbers of IFNγ-producing CD4+ and CD8+ T cells were seen
212
at early time points (2 weeks post-infection), but the difference became non-significant at later time points
213
(Suppl. Fig. 5), consistent with the hypothesis that a fast innate control of bacterial load results in a swift
214
contraction of the TB-specific Th1 and Tc1 cellular response after sapM::Tn BCG vaccination. The sapM::Tn
215
BCG vaccine thus behaves like a rapidly (though not completely) controlled bacterial infection, contrary to
216
WT BCG, which expands and causes a more protracted infection at the vaccination site.
217
3. Discussion
218
Our former research demonstrated that an M. bovis BCG sapM::Tn mutant led to enhanced long-term
219
survival of vaccinated mice challenged with TB [13]. Further development of this TB vaccine candidate
220
requires further characterization of the mutant and its safety, which was the main purpose of the extensive
221
work presented in this study.
222
We have performed a whole genome resequencing-based variant analysis of this sapM::Tn BCG mutant, as
223
well as of its parent BCG strain [26], using a shotgun Illumina sequencing approach. Very few polymorphisms
224
could be detected compared to the M. bovis BCG Pasteur reference genome, most of which were already
225
present in the streptomycin-resistant Pasteur derivative in which we designed our transposon mutant. A
226
frame-shift mutation had inactivated the gene coding for FadD26, a key enzyme in the biosynthesis pathway
227
of the PDIM class of virulence lipids that are involved in hiding M. tb’s own PAMPs from the host’s innate
228
immune system [29]. Knocking out this gene in M. tb severely impairs the pathogen’s ability to survive in vivo
229
[29,30]. For this reason, the fadD26 gene is deleted in MTBVAC, the first live-attenuated M. tb vaccine [37].
230
Chen et al. compared PDIM/PGL production in 12 different M. bovis BCG substrains and found that while
12
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
231
most were PDIM-positive (including BCG Pasteur), three of these strains, BCG Japan (or Tokyo), Moreau and
232
Glaxo, do not produce this class of lipids [38]. The authors further reported that there was a correlation
233
between the virulence that was associated with the BCG substrains and their lipid profile: the on average
234
more virulent, PDIMs/PGLs-producers and the on average less virulent, PDIMs/PGLs non-producers. In a later
235
study it was shown that the PDIM-defect in BCG Moreau was due to a deletion in the FadD26 and the directly
236
downstream ppsA gene in the same operon, which is also involved in the PDIM/PGL lipid biosynthesis
237
pathway [39]. The authors reported that this locus was intact in BCG Japan and Glaxo. Naka et al later showed
238
that BCG Tokyo 172 is actually divided into two subpopulations, in which type II, but not type I, has a
239
frameshift mutation in the ppsA gene and thus does not produce PDIM lipids [40]. It is unclear whether this
240
strain is the same as the Japan strain that was earlier reported to lack PDIMs [38]. As the Japan/Tokyo,
241
Moreau and Glaxo substrains are not derived from one another [41], and as we have here detected a fadD26
242
frameshift in a substrain derived from BCG Pasteur, these independent mutations in genes coding for key
243
enzymes in the PDIM/PGL lipid biosynthesis pathway indicate that there may be a selective force in favor of
244
losing this class of lipids during in vitro BCG cultivation. We also conclude that, as our strain has this
245
background fadD26 mutation, the sapM::Tn vaccine strain in our studies is in fact a fadD26/sapM::Tn double
246
mutant, which is important knowledge for further development, as it shows that the sapM mutation
247
improves vaccine efficacy even in this fadD26 background, as is currently being used for safety enhancement
248
of the MTBVAC lead clinical TB vaccine candidate. In addition to our sapM::Tn BCG mutant, this FadD26
249
mutation will almost certainly also be present in the Zmp1 BCG mutant candidate vaccine that was described
250
in [42], since it is also derived from the M. bovis BCG Pasteur 1721 strain.
251
By comparative transcriptome profiling between WT BCG and sapM::Tn BCG, we observed an almost
252
complete absence of sapM transcript in the mutant strain, together with a significant upregulation of sapM’s
253
upstream gene upp. By RNAseq analysis, we further showed that the global transcriptional landscape in the
254
sapM::Tn BCG mutant strain is not significantly altered compared to the WT BCG.
255
The most promising strategy for improved TB vaccines currently is the development of an improved
13
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
256
attenuated live vaccine, either used alone or in combination with subunit booster vaccines [6]. Furthermore,
257
re-vaccination with BCG in adolescence has recently shown promise as well [43]. An improved version of the
258
BCG vaccine could thus be valuable in this context too. However, working with live vaccines encompasses
259
potential safety issues, especially since a high incidence of HIV is found in areas where TB is endemic. A novel
260
vaccine candidate should therefore be at least as safe as BCG. Safety of the sapM::Tn BCG strain was tested
261
in immunocompromised SCID mice, in comparison with the WT BCG. Survival time upon infection of the mice
262
was similar for both strains, at both high and low infectious doses, demonstrating comparable safety of the
263
sapM::Tn BCG strain and WT BCG.
264
Due to the lack of natural infection-induced protection, the type of immune response that is crucial in
265
preventing M.tb infection and TB, and which therefore should be induced by vaccination, remains largely
266
unknown. At present, it is believed that a particular balance between differentially polarized adaptive
267
immune responses is crucial to overcome a TB infection. M. tb as well as M. bovis, from which BCG derives,
268
have evolved immunomodulatory mechanisms to perturb this balance to its own benefit [44,45]. We and
269
others propose that an improved, live attenuated vaccine will need to be engineered to take down these
270
immunomodulatory virulence factors from the pathogenic parental Mycobacteria from which these vaccines
271
(incl. BCG) have been derived, and we propose that the SapM secreted phosphatase is such a factor. Our
272
results demonstrate a better innate control of sapM::Tn BCG vaccine bacteria as compared to WT BCG upon
273
vaccination, correlating with a faster influx of iDCs in lymphoid organs draining the vaccination site, and more
274
modest primary expansion of TB-antigen-specific IFNγ-producing CD4 + and CD8+ T cells. Both WT BCG and
275
sapM::Tn BCG have equivalent growth characteristics during in vitro cultivation (Suppl. Fig. 6), excluding that
276
mere growth rate differences could have caused any difference in abovementioned phenotypes. Since
277
complementation of sapM expression abolished the improved growth control of the sapM::Tn BCG mutant
278
upon infection, the observed phenotypes are specifically due to the reduced expression of sapM. To
279
conclude, all of this shows that the sapM mutated BCG vaccine behaves like a more immediately effectively
280
controlled bacterial infection as compared to WT BCG. Importantly, in the context of viral infections, it has
281
been well established that a protracted fight of the immune system to a chronic infection yields poorly
14
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
282
efficacious memory due to poor generation of central memory T cells [46]. Possibly, the same holds true for
283
chronic bacterial intracellular infections, of which BCG vaccination is an example. Together with the
284
comparable safety of the SapM-mutated BCG and WT BCG in immunodeficient mice, and the fact that the
285
sapM::Tn BCG vaccine is presently one of the few priming vaccines for which enhanced long-term survival of
286
TB-infected animals has been demonstrated [13], these results warrant inclusion of the SapM inactivation
287
strategy in future generations of live attenuated TB vaccines.
288
4. Materials and Methods
289
4.1 Mycobacterial strains and media
290
The streptomycin resistant M. bovis BCG Pasteur strain 1721 [26] (RpsL, K43R; a gift of Dr. P. Sander, Institute
291
for Medical Microbiology, Zürich) and its sapM transposon insertion mutant [13] were grown in shaking
292
culture flasks in Middlebrook 7H9 broth (Difco) supplemented with 0.05% Tween80 and Middlebrook OADC
293
(Becton Dickinson) when grown in liquid culture. Difco Middlebrook 7H10 agar was used for growth on solid
294
culture. For the SDS-PAGE and western blotting experiment, M. bovis BCG cultures were grown in 7H9
295
medium without OADC, but supplemented with 0.2% glucose, 0.2% glycerol and 0.05% of Tween80, because
296
the large quantity of albumin in the OADC supplement interferes with protein electrophoresis. All strains
297
were frozen as 1 mL cultures at OD600 1 in 20% glycerol (final concentration). For all in vitro and in vivo
298
infections described in this paper, cells were started from a glycerol stock, grown in shaking culture flasks
299
(OD600 always ≤1) as described above, subcultured once and further grown until OD600~0.8 after which cells
300
were collected and prepared for infection. For each experiment, we confirmed that we immunized with the
301
same number of viable CFU by CFU plating of the inoculum.
302
4.2 Whole genome Illumina resequencing and data analysis
303
Genomic DNA of the M. bovis BCG strain 1721 and its sapM::Tn BCG mutant (Tn-7432) was prepared [47]
304
and used for an Illumina library prep (Nextera XT DNA Library Preparation kit). The libraries were sequenced
305
on an Illumina MiSeq instrument (2 x 150 bp reads), with an average 80x coverage per genome. The library
15
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
306
prep and sequencing was performed by the VIB Nucleomics Core facility (www.nucleomics.be). The data was
307
analyzed by the VIB Nucleomics Core facility and the VIB Bioinformatics Core facility (www.bits.vib.be) using
308
CLC Genomics Workbench (CLC-GWB [48], http://www.clcbio.com/products/clc-genomics-workbench/). All
309
reads were processed using the standard CLC-GWB settings. The reads from the sapM::Tn BCG strain were
310
used for de-novo assembly in order to locate the inserted transposon. To this end, the phiMycoMarT7
311
transposon sequence (GenBank AF411123.1) was blasted against all de-novo contigs (n = 201). Using this
312
information, a modified reference sequence was built from the M. bovis BCG Pasteur str. 1173P2 reference
313
(NCBI NC_008769.1) by introducing the phiMycoMar T7 transposon sequence. All reads from both strains
314
were then mapped to their corresponding reference sequences using the CLC-GWB default command
315
settings (mismatch cost 2, insertion cost 3, deletion cost 3, length fraction 0.5, similarity fraction 0.8). A
316
probabilistic variant analysis as implemented in the CLC-Genomics Workbench package [48] was then
317
performed (min coverage 10, 90% variant probability) to obtain the variants listed in Table 1.
318
To confirm these variants, PCR primers were designed flanking the mutations, after which the regions were
319
amplified (Phusion polymerase), purified (AMPure XP beads) and Sanger sequenced (VIB Genetics Service
320
Facility (http://www.vibgeneticservicefacility.be) with nested primers (see table below). PCR conditions were
321
as follows: 98°C for 3 minutes; 25 cycles of denaturation (98°C for 20 seconds), annealing (69°C for 20
322
seconds) and extension (72°C for 1 minute); 72°C for 5 minutes. The primer details are given in Suppl. Table
323
2.
324
4.3 RT-qPCR analysis
325
10 mL of M. bovis BCG cultures (grown in standard 7H9 medium until an OD600 of 0.8 – 1.0) were centrifuged
326
and the pellets were washed once with 0.5% Tween80 solution in PBS. The pellet was then resuspended in
327
500 µl of RLT buffer (RNeasy Mini Kit, Qiagen; supplemented with β-ME). The cells were disrupted with glass
328
beads in Retsch MM2000 bead beater at 4°C in screw-cap tubes (pre-baked at 150°C). After centrifugation (2
329
min, 13,000 rpm, 4°C), the supernatant was transferred to a fresh eppendorf tube. To recover the lysate
330
trapped in between the beads, 800 µl of chloroform was added to the beads and centrifuged, after which the
16
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
331
upper phase was transferred to the same eppendorf tube as before. Then, 1 volume of Acid
332
Phenol/Chloroform (Ambion) was added, incubated for 2 minutes and centrifuged (5 min, 13,000 rpm, 4°C).
333
The upper aqueous phase was transferred to a fresh eppendorf tube and this last step was repeated once.
334
An equal volume of 70% ethanol was added and the sample was transferred to an RNeasy spin column
335
(RNeasy Mini Kit, Qiagen). The kit manufacturer’s instructions were followed to purify the RNA. After elution
336
in RNase-free water (30 µl), an extra DNase digestion was performed with DNaseI (10 U of enzyme, 50 µl
337
reaction volume). Then an extra clean-up step was performed with the RNeasy Mini Kit (Qiagen). Finally, the
338
RNA concentration was determined on a Nanodrop instrument. Typically, RNA yields of 2-3 µg were obtained
339
with this method.
340
cDNA was prepared from 1 μg of DNase-treated RNA using the iScript Synthesis Kit (BioRad) and a control
341
reaction lacking reverse transcriptase was included for each sample. The RT-PCR program was as follows: 10
342
minutes at 25°C, 30 minutes at 42°C, 5 minutes at 85°C, and cooling down to 12°C.
343
Real time quantitative PCR was done on a LightCycler 480 (Roche Diagnostics) using the SensiFast SYBR-
344
NoRox kit (BioLine), in triplicate for each cDNA sample, on a 384-multiwell plate, with 1 ng of cDNA in a total
345
volume of 10 μL. Primers were used at a final concentration of 10 μM. All primer pairs were generated using
346
Primer3Plus (http://www.bioinformatics.nl/cgi-bin/primer3plus/primer3plus.cgi) as described [49]. All gene
347
expression values were normalized using the geometric mean of the GroEL gene and 16S rRNA.
348
Determination of amplification efficiencies and conversion of raw Cq values to normalized relative quantities
349
(NRQ) was performed using the qbasePLUS software (Biogazelle). Statistical analysis of the NRQs was done
350
with the Prism 6 software package using a two-tailed t-test. Primer details are given in Suppl. Table 2.
351
4.4 RNA-Seq analysis
352
M. bovis BCG cultures (sapM::Tn BCG mutant and WT BCG 1721) were grown in standard 7H9 medium in
353
triplicates until an OD600 of 0.8 – 1.0 and RNA was prepared as described for the RT-qPCR analysis. This RNA
354
was then depleted of ribosomal RNA (Ribo-Zero rRNA removal kit for gram-positive bacteria) and the
355
remaining RNA was prepared for Illumina sequencing (TruSeq stranded total RNA preparation kit). This library
17
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
356
preparation takes into account the strandedness of the RNA and is able to distinguish between sense and
357
antisense transcripts. The libraries were sequenced on an Illumina NextSeq500 (75 bp single-end). The library
358
preparation and sequencing was performed by the VIB Nucleomics Core facility (www.nucleomics.be). The
359
data was analyzed using the standard RNA-seq analysis workflow in CLC Genomics Workbench v7 (CLC-GWB
360
[48], http://www.clcbio.com/products/clc-genomics-workbench/). In short, raw sequencing reads were
361
quality trimmed and mapped to the reference genome (M. bovis BCG Pasteur str. 1173P2), taking into
362
account the read directionality. The total number of unique reads per gene (averaged over the triplicate
363
samples) of both the WT BCG and the sapM::Tn BCG strain were then compared using the EdgeR statistical
364
test [50,51].
365
4.5 Recombinant SapM production and anti-SapM antibody preparation
366
To express SapM in E. coli, we constructed an expression vector pLSPelB-SapM mature-His, being pLSAH36
367
carrying a PelB signal sequence, mature SapM without signal sequence and a C-terminal His tag. This was
368
transformed to the E. Coli BL21+pICA2 expression strain. E. coli was grown in LB medium supplemented with
369
ampicillin and kanamycin at 28°C. Expression was induced with 1 mM IPTG overnight (ON) at 18°C. Bacteria
370
were centrifuged and the pellet was sonicated in sonication buffer (50 mM Tris, 100 mM NaCl, 1 mM PMSF,
371
pH8). After centrifugation at 18000 rpm for 1h, supernatant was removed and inclusion bodies were
372
extracted in extraction buffer (20 mM NaH2PO4, 0,5 M NaCl, 5 mM β-mercaptoethanol, 6 M guanidine
373
hydrochloride, 5 mM imidazole, pH 7.5) ON at 4°C. Denatured protein was purified using nickel sepharose
374
and eluted in elution buffer (20 mM NaH2PO4, 50 mM NaCl, 5 mM β-mercaptoethanol, 8 M guanidine
375
hydrochloride, 400 mM imidazole, pH 7.5) and a final desalting was performed on a sephadex G25. The
376
purification was performed by the VIB Protein Core facility. The rabbit polyclonal anti-SapM antibody was
377
generated by immunization of 2 rabbits (CER groupe, Belgium) with purified SapM protein (625µg/mL, in the
378
text named anti-SapM; denatured, full-length protein produced in E. coli as described above). Immunizations
379
and bleedings were done following the standard CER protocol. In brief, rabbits were subcutaneously injected
380
on days 0, 14, 28 and 56 with 500µL antigen/500µL adjuvant. Bleedings were performed on day 0
18
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
381
(preimmune 2mL), day 38 (small test sample 2mL), day 66 (large bleed 20+2mL) and day 80 (final bleeding).
382
Sample from the final bleeding of rabbit 2 is used in the experiments described here. The rabbit polyclonal
383
anti-SapM-C-term antibody (Anti-PepC) was raised against a synthetic C-terminal peptide (peptide synthesis
384
(CYATNAPPITDIWGD, terminal modification: amidation (C-terminal) and conjugation to KLH) and
385
immunizations performed by GenScript). Immunizations and bleedings were done following the standard
386
GenScript protocol. In brief, 2 rabbits were subcutaneously injected on days 0 and day 14. First bleedings
387
were performed on day 0 (pre-immune, 2mL) and 21 (1mL). A third immunization was performed on day 35.
388
A second bleeding was done on day 42 (1mL). A fourth immunization was done on day 56. Final bleeding was
389
done on day 63. Only pre-immune serum and final anti-sera was provided by Genscript.
390
4.6 SDS-PAGE and western blotting
391
M. bovis BCG cultures were grown in 7H9 medium without OADC (as the albumin in the OADC supplement
392
would otherwise interfere on SDS-PAGE), but supplemented with 0.2% glucose, 0.2% glycerol and 0.05% of
393
Tween80. The supernatant samples were precipitated by first adding sodium deoxycholate (DOC, 0.1% final
394
concentration), incubating 15 min on ice and then adding trichloroacetic acid (TCA, 7.7% final concentration)
395
followed by 30 min of incubation on ice. The samples were centrifuged and the pellet was washed twice with
396
acetone and dissolved again in PBS with Laemmli loading dye. Precipitated protein samples corresponding to
397
200 µl of culture medium were analyzed by SDS-PAGE and subsequent western blotting. The membranes
398
were blocked overnight (4°C) with PBST (PBS + 0.05% Tween-20) containing 5% milk powder. To visualize the
399
SapM protein, the blots were incubated for 1h at room temperature with a rabbit anti-SapM polyclonal
400
antibody (1:5,000 dilution in PBST), washed 3 times with PBST and then incubated for 1h with a secondary
401
goat anti-rabbit DyLight 800 conjugated antibody (30 ng/ml in PBST). After the final washing steps, the
402
membranes were scanned with a LI-COR Odyssey system.
403
4.7 ELISA
404
M. bovis BCG culture supernatant samples (grown in 7H9; 50 µl) or control samples (0.25 µg/ml of purified
405
SapM protein) were diluted 1:2 in 100 mM of carbonate buffer (3.03 g of Na2CO3 and 6 g of NaHCO3 in 1 L of
19
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
406
water; pH 9.6) and coated on 96-well maxisorp plates (Nunc) (overnight at 4°C). The plate was then washed
407
three times with PBST (PBS + 0.05% Tween-20). The plate was blocked with 1% of BSA in PBS (2 hours at room
408
temperature (RT)). After plate washing, the rabbit polyclonal anti-SapM or anti-PepC antibody was added
409
(diluted 1:5000 in PBS with 0.1% Tween-20 and 0.1% goat serum) (1 hour at RT). After washing, the plate was
410
incubated with the secondary HRP-labeled donkey anti-rabbit antibody (GE healthcare) (diluted 1:5000 in
411
PBS with 1% BSA) (1 hour at RT). After washing, 100 µl of the TMB substrate was added to the plate (Becton
412
Dickinson), incubated at RT for 30 minutes, after which 50 µl of 2N H2SO4 was added to stop the reaction.
413
The optical density at 450 nm was then determined with a bio-rad plate reader (wavelength correction at
414
655 nm).
415
4.8 Phosphatase assay
416
A 150 µl reaction mixture, containing 50 µl of M. bovis BCG culture supernatant samples, 85 µl of sodium
417
acetate (0.2 M; pH 7.0) and 15 µl of para-nitrophenylphosphate (pNPP, 50 mM), was incubated for 6 hours
418
or overnight at 37°C. Finally, 50 µl of stop buffer (1 M of Na2CO3) was added and the optical density at 415
419
nm was determined with a bio-rad plate reader.
420
4.9 Laboratory animals
421
Female C57BL/6J mice (Janvier), C57BL/6J x Balb/c mice (F1, Harlan) and CB-17 SCID mice (Charles River)
422
were housed under specific pathogen-free conditions in micro-isolator units. At the beginning of the
423
experiment, the mice were 7–8 weeks old. All experiments were approved by and performed according to
424
the guidelines of the ethical committee of Ghent University, Belgium.
425
4.10
426
Eight-week-old female CB-17 SCID mice (Charles River Laboratories, 7 or 8 mice/group as indicated) were
427
immunized intravenously via tail vein injection with 100 μL (3x10 6 versus 3x107 cfu) of WT BCG Pasteur or
428
sapM::Tn BCG mutant (diluted in PBS). The control group received PBS (pH 7.5). The animals were observed
429
and weighed initially on a ~2-weekly basis and from day 82 post-infection on, ~every 2 days (until day 201
SCID safety study
20
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
430
post-infection). The differences of time to survival between the different groups were compared using log-
431
rank test. Weight loss of 20% of initial body weight was set as the ethical endpoint.
432
4.11
433
Bone marrow cells from C57BL/6J mice were harvested and differentiated to macrophages in DMEM
434
containing 20% of L929-cell supernatant, 10% of heat-inactivated FCS, 0.4 mM sodiumpyruvate, 50 µM β-
435
mercaptoethanol and 0.1 mM MEM non-essential amino acids for 7 days. Cells were cultured in antibiotic-
436
free medium at all times. To analyze the survival/replication of M. bovis BCG following infection of
437
macrophages, bone marrow-derived macrophages (BM-DMs, differentiated as described in [52]) were either
438
left untreated or infected with WT BCG versus sapM::Tn BCG (MOI 10:1) in infection medium (DMEM +
439
supplements described earlier) for 4 hours. Cells were then washed 2x with pre-warmed PBS and 1x with pre-
440
warmed medium to remove extracellular bacteria. Fresh medium was added and cells were further incubated
441
for the indicated time points. BM-DMs were lysed in cell lysis buffer (0.05% v/v SDS, 200mM NaCl, 10mM
442
Tris-HCl pH7.5, 5mM EDTA, 10% glycerol) at the indicated times and serial dilutions were plated onto
443
Middlebrook 7H10 agar plates followed by incubation at 37°C.
444
4.12
445
Three experiments were performed. At the age of 8 weeks, female C57BL/6J x Balb/c (F1, Harlan) mice were
446
infected s.c. (at the base of the tail) with 2x106 CFU of WT BCG or sapM::Tn BCG mutant in 100 µl of PBS. At
447
day 8, 14 and 28 post-infection, 10 mice/group were sacrificed by cervical dislocation; inguinal and brachial
448
LNs were removed aseptically and homogenized in PBS-0.5% Tween 20 + EDTA-free complete protease
449
inhibitor (Roche). Neat, 1/10, 1/100 and 1/1000 dilutions were made of which 50µL was plated in duplicate
450
on 7H10-OADC agar and the colonies were counted after 21 days. The CFU counts were calculated from the
451
highest dilution showing distinct colonies.
452
In another experiment, at the age of 8 weeks, female C57BL/6J x Balb/c (F1, Harlan) mice were infected
453
intravenously (i.v.) with 2x106 CFU WT BCG or sapM::Tn BCG mutant in 200 µl of PBS. At 24h and 3 weeks
Cell culture and in vitro infection
Vaccination with M. bovis BCG and analysis of in vivo BCG CFU dynamics
21
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
454
post-infection, 8 mice/group were sacrificed; spleens and lungs were removed aseptically, homogenized in
455
PBS-0.5% Tween 20 + EDTA-free complete protease inhibitor (Roche) and analyzed for CFU counts as above.
456
Again, neat, 1/10, 1/100 and 1/1000 (lung) and neat, 1/10, 1/100, 1/1000, 1/10 000, 1/100 000 (spleen)
457
dilutions were made of which 50µL was plated in duplicate on 7H10-OADC agar. Additionally, 6h, 24h and
458
48h post-infection, blood was taken retro-orbitally and collected in BD microtainer SST tubes for serum
459
preparation, which was frozen at -20°C until assay. Serum cytokines were determined by bioplex (BioPlex Pro
460
Cytokine Express Assay, Biorad).
461
In at third experiment, at the age of 8 weeks, female C57BL/6J (Janvier) mice were infected s.c. (at the base
462
of the tail) with 2x106 CFU WT BCG or sapM::Tn BCG mutant in 100 µl PBS. At day7, 14 and 28 post-infection,
463
6 mice/group were sacrificed; inguinal and brachial LNs were removed aseptically and analyzed for CFU
464
counts as above.
465
4.13
466
To complement the sapM::Tn BCG mutant strain with sapM gene, we constructed an integration plasmid
467
carrying sapM as follows: the SapM operon was amplified by PCR using primers SapMPo-XbaI
468
(CACTCTAGACAGTGGGTGGTCAACGACA) and Rv3311HindIII (GTTAAGCTTCCGGCTGACGGTGCCTATTC) from
469
gDNA of BCG Pasteur. The PCR fragment was cloned in pMV261Hyg (kindly provided by Prof. E. Rubin) via
470
XbaI and HindIII to form pMV261HygSapMoperon. The vector pMV261HygSapMoperon was digested with
471
XbaI and ApaLI and the integrative vector pMV306Kan (kindly provided by Prof. W.R. Jacobs jr.) was digested
472
with NheI and ApaLI, after which the appropriate fragments were purified from gel and ligated together to
473
form pMV306HygSapMoperon, an integrating vector with the SapM operon (Plasmid map in Suppl. Fig. 7).
474
The plasmid pMV306HygSapMoperon was electroporated into the sapM::Tn BCG mutant strain [53].
475
Hygromycin resistant colonies were selected and surviving clones were tested on a phosphatase assay.
476
4.14
477
C57BL/6J mice were immunized s.c. at the base of the tail with WT BCG or sapM::Tn BCG mutant, either with
478
a high dose (2x106 cfu, 5-10 mice/group) or a low dose (2x105 cfu, 10 mice/group). Seven and 14 days later,
22
Complementation sapM::Tn BCG mutant
Vaccination with M. bovis BCG and analysis of DC phenotype
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
479
the mice were sacrificed, and inguinal LNs, brachial LNs and spleens were isolated. Cells were prepared by
480
crushing the organs on a 70µm cell strainer and labelled with an optimized DC-analysis antibody panel and
481
analyzed by flow cytometry. Fluorochrome-conjugated mAbs specific for mouse CD11c (clone N418,
482
ebioscience), CD8 (clone 53-6.7, BD), CD40 (clone 1C10, ebioscience), CD80 (clone 16-10A1, Biolegend), Ly-
483
6C (clone AL-21, BD), MHCII (clone M5/114.15.2, ebioscience), CD103 (clone 2E7, ebiosciene), F4/80 (clone
484
BM8, ebioscience), CD11b (clone M1/70.15, Life Technologies), Ly-6G (clone RB6-8C5, ebioscience) were
485
used. Dilutions of the antibodies are given in Suppl. Table 3 and gating strategy is given in Suppl. Fig. 8.
486
4.15
487
C57BL/6J mice were vaccinated s.c. at the base of the tail with either a high dose (2x10 6 cfu) or a low dose
488
(2x105 cfu, in 100µL of PBS) WT BCG or sapM::Tn BCG mutant. 14 days after infection, mice were killed by
489
cervical dislocation and inguinal + brachial LNs and spleens were removed aseptically. Cells were prepared
490
by crushing the organs on a 70 µm cell strainer and adjusted to a concentration of 1x106 (LN) or 2x106 (spleen)
491
cells/ mL in either flat-bottom 96-well microwell plates (Bioplex/ELISA) or round-bottom plates (flow
492
cytometry) (Nunc), in RPMI 1640 medium, supplemented with 10% fetal calf serum, L-glutamine (0.03%), 0,4
493
mM sodiumpyruvate, 0,1 mM non-essential amino acids and 50 μM β-mercaptoethanol. Cells were
494
restimulated with purified native Ag85A (Rv3804c) from M. tb, strain H37Rv (15 µg/mL in PBS, obtained
495
through BEI Resources, NIAID, NIH), purified protein derivative (PPD, 10 μg/mL, produced at Scientific
496
Institute for Public Health, Brussels, Belgium) and anti-CD3/28 (10 µg/mL, BD) in a volume of 100 μL added
497
to 100 μL of cell suspension. For surface and intracellular staining, cells were incubated at 37°C in a humidified
498
CO2 incubator, and after 1h, brefeldin was added (10 µg/mL-Sigma Aldrich) for another 5 hours. LNs and
499
splenocytes were subsequently labelled with an optimized antibody panel for T cell phenotyping and
500
analyzed by flow cytometry. Fluorochrome-conjugated mAbs specific for mouse CD3 (clone 17A2, BD), NK1.1
501
(clone PK136, BD), CD4 (clone RM4-5, Invitrogen), CD8 (clone 53-6.7, BD), CD335 (clone 29A1.4, BD), TCRγδ
502
(clone GL-3, ebioscience), IL17 (clone TC11-18H10, BD), IFNγ (clone XMG1.2, BD) and TNF (clone MP6-XT22,
503
ebioscience) were used. Fixable viability dye eFluor 780 (eBioscience) was used as a live/dead marker.
Vaccination with M. bovis BCG for T cell phenotyping and antigen-specific cytokine determination
23
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
504
Dilutions of the antibodies and gating strategy are given in Suppl. Materials and Methods. For determination
505
of cytokine levels in the supernatant, the latter was harvested after 48h post-stimulation. Supernatants were
506
frozen at -20°C until assay. Cytokines were determined by bioplex (BioPlex Pro Cytokine Express Assay,
507
Biorad) and ELISA (mouse IFNγ duoset Elisa, R&D systems) as indicated in the text.
508
4.16
509
Cells were first blocked by adding Fc block (purified rat anti-mouse CD16/CD32 monoclonal Ab, clone 2.4G2,
510
BD, 0.1µg/sample) and incubating for 15 min at 4°C. Staining is performed for 20 min at 4 °C in the dark. For
511
intracellular staining, cells were incubated for 5 h with 10 µg/mL brefeldin A (10 µg/mL- Sigma Aldrich) before
512
fixation and permeabilization (Fixation-Permeabilization concentrate, Permeabilization buffer, ebioscience).
513
In all analyses, following doublet exclusion, live cells were identified using a fixable viability dye (Molecular
514
Probe, Life Technologies). Data were acquired on an LSR II equipped with three lasers (488, 635, 405nm)
515
(APC panel) or Fortessa with 4 lasers (405, 488, 561, 633 nm) (T cell panel) (BD Biosciences) and analyzed
516
using FlowJo software (Tree Star, Ashland, OR).
517
4.17
518
Results are presented as means ± standard error of the mean (SEM) unless otherwise stated and groups
519
were compared using statistical tests as mentioned in the text, using Prism Software (GraphPad Software,
520
San Diego, CA).
521
5. Acknowledgements
522
This work was supported by an ERC Consolidator grant to N.C. (GlycoTarget; 616966) and a GOA grant. N.F.
523
and D.V. were postdoctoral fellows and K.V. and K.B. were predoctoral fellows of FWO (Fonds
524
Wetenschappelijk Onderzoek-Vlaanderen). We thank Dr. Peter Sander (Institute for Medical Microbiology,
525
Faculty of Medicine, University of Zurich) for providing us with the M. bovis BCG strain (strain 1721, RpsL,
526
K43R), Dr. S. Alonso (Dept, Microbiology, National University of Singapore) for providing us the optimal
527
protocol for electroporation of M. bovis BCG and Dr. K. Huygen (WIV, Brussels) for providing us with the PPD.
24
Flow cytometry analysis
Statistical analysis
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
528
We thank Prof. Dr. E. Rubin (Dept, Immunology and Infectious diseases, Harvard School of Public Health,
529
Boston) for providing us the pMV261Hyg vector and Prof. W.R. Jacobs Jr (Albert Einstein College of Medicine,
530
New York) for providing us the pMV306Kan vector. We thank the VIB Nucleomics core (www.nucleomics.be)
531
and Genomics Core (UZ – K.U. Leuven, gc.uzleuven.be) facilities for the Illumina MiSeq and HiSeq sequencing
532
services. We thank Jannick Leoen from the VIB Protein Core facility (https://corefacilities.vib.be/psf) for
533
purification of SapM protein. The following reagent was obtained through BEI Resources, NIAID, NIH: Ag85
534
Complex, Purified Native Protein from Mycobacterium tuberculosis, Strain H37Rv, NR-14855.
535
N.C and N.F. are named inventors on a patent covering the use of SapM mutation as a vaccine improvement
536
strategy.
537
6. Author contributions
538
NF coordinated the project, designed and performed experiments, analyzed and interpreted data and wrote
539
the manuscript. KV analyzed genome sequencing and RNAseq experiments, performed RT-qPCR experiments
540
and co-wrote the manuscript. EH designed and performed in vivo experiments, analyzed and interpreted
541
data. EP performed all in vitro experiments and helped in performing in vivo experiments. DV and KB helped
542
in performing in vivo experiments. PT and AV made the sapM::Tn:complementation mutant. NC coordinated
543
the project, interpreted data and co-wrote the manuscript.
544
25
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
545
7. References
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
WHO | Global tuberculosis report 2018. WHO n.d.
http://www.who.int/tb/publications/global_report/en/.
Nieuwenhuizen NE, Kaufmann SHE. Next-Generation Vaccines Based on Bacille Calmette–Guérin.
Front Immunol 2018;9. doi:10.3389/fimmu.2018.00121.
Rodrigues LC, Diwan VK, Wheeler JG. Protective effect of BCG against tuberculous meningitis and
miliary tuberculosis: a meta-analysis. Int J Epidemiol 1993;22:1154–8.
Trunz BB, Fine P, Dye C. Effect of BCG vaccination on childhood tuberculous meningitis and miliary
tuberculosis worldwide: a meta-analysis and assessment of cost-effectiveness. Lancet Lond Engl
2006;367:1173–80. doi:10.1016/S0140-6736(06)68507-3.
Colditz GA, Brewer TF, Berkey CS, Wilson ME, Burdick E, Fineberg HV, et al. Efficacy of BCG vaccine in
the prevention of tuberculosis. Meta-analysis of the published literature. JAMA 1994;271:698–702.
Hatherill M, Tait D, McShane H. Clinical Testing of Tuberculosis Vaccine Candidates. Microbiol Spectr
2016;4. doi:10.1128/microbiolspec.TBTB2-0015-2016.
Triccas JA, Counoupas C. Novel vaccination approaches to prevent tuberculosis in children.
Pneumonia 2016;8. doi:10.1186/s41479-016-0020-z.
Horwitz MA, Harth G. A New Vaccine against Tuberculosis Affords Greater Survival after Challenge
than the Current Vaccine in the Guinea Pig Model of Pulmonary Tuberculosis. Infect Immun
2003;71:1672–9. doi:10.1128/IAI.71.4.1672-1679.2003.
Martin C, Williams A, Hernandez-Pando R, Cardona PJ, Gormley E, Bordat Y, et al. The live
Mycobacterium tuberculosis phoP mutant strain is more attenuated than BCG and confers protective
immunity against tuberculosis in mice and guinea pigs. Vaccine 2006;24:3408–19.
doi:10.1016/j.vaccine.2006.03.017.
Sun R, Skeiky YAW, Izzo A, Dheenadhayalan V, Imam Z, Penn E, et al. Novel recombinant BCG
expressing perfringolysin O and the over-expression of key immunodominant antigens; pre-clinical
characterization, safety and protection against challenge with Mycobacterium tuberculosis. Vaccine
2009;27:4412–23. doi:10.1016/j.vaccine.2009.05.048.
Brandt L, Skeiky YAW, Alderson MR, Lobet Y, Dalemans W, Turner OC, et al. The Protective Effect of
the Mycobacterium bovis BCG Vaccine Is Increased by Coadministration with the Mycobacterium
tuberculosis 72-Kilodalton Fusion Polyprotein Mtb72F in M. tuberculosis-Infected Guinea Pigs. Infect
Immun 2004;72:6622–32. doi:10.1128/IAI.72.11.6622-6632.2004.
Bertholet S, Ireton GC, Ordway DJ, Windish HP, Pine SO, Kahn M, et al. A Defined Tuberculosis Vaccine
Candidate Boosts BCG and Protects Against Multidrug Resistant Mycobacterium tuberculosis. Sci
Transl Med 2010;2:53ra74. doi:10.1126/scitranslmed.3001094.
Festjens N, Bogaert P, Batni A, Houthuys E, Plets E, Vanderschaeghe D, et al. Disruption of the SapM
locus in Mycobacterium bovis BCG improves its protective efficacy as a vaccine against M.
tuberculosis. EMBO Mol Med 2011;3:222–34. doi:10.1002/emmm.201000125.
Puri RV, Reddy PV, Tyagi AK. Secreted acid phosphatase (SapM) of Mycobacterium tuberculosis is
indispensable for arresting phagosomal maturation and growth of the pathogen in guinea pig tissues.
PloS One 2013;8:e70514. doi:10.1371/journal.pone.0070514.
Bahal RK, Mathur S, Chauhan P, Tyagi AK. An attenuated quadruple gene mutant of Mycobacterium
tuberculosis imparts protection against tuberculosis in guinea pigs. Biol Open 2017;7.
doi:10.1242/bio.029546.
Zulauf KE, Sullivan JT, Braunstein M. The SecA2 pathway of Mycobacterium tuberculosis exports
effectors that work in concert to arrest phagosome and autophagosome maturation. PLoS Pathog
2018;14. doi:10.1371/journal.ppat.1007011.
Saikolappan S, Estrella J, Sasindran SJ, Khan A, Armitige LY, Jagannath C, et al. The fbpA/sapM Double
Knock Out Strain of Mycobacterium tuberculosis Is Highly Attenuated and Immunogenic in
Macrophages. PLoS ONE 2012;7. doi:10.1371/journal.pone.0036198.
26
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
[18] Nunes-Alves C, Booty MG, Carpenter SM, Jayaraman P, Rothchild AC, Behar SM. In search of a new
paradigm for protective immunity to TB. Nat Rev Microbiol 2014;12:289–99.
doi:10.1038/nrmicro3230.
[19] Chin KL, Anis FZ, Sarmiento ME, Norazmi MN, Acosta A. Role of Interferons in the Development of
Diagnostics, Vaccines, and Therapy for Tuberculosis. J Immunol Res 2017;2017.
doi:10.1155/2017/5212910.
[20] Cowley SC, Elkins KL. CD4+ T Cells Mediate IFN-γ-Independent Control of Mycobacterium tuberculosis
Infection Both In Vitro and In Vivo. J Immunol 2003;171:4689–99. doi:10.4049/jimmunol.171.9.4689.
[21] Gallegos AM, van Heijst JWJ, Samstein M, Su X, Pamer EG, Glickman MS. A Gamma Interferon
Independent Mechanism of CD4 T Cell Mediated Control of M. tuberculosis Infection in vivo. PLoS
Pathog 2011;7. doi:10.1371/journal.ppat.1002052.
[22] Orr MT, Windish HP, Beebe EA, Argilla D, Huang P-WD, Reese VA, et al. Interferon γ and Tumor
Necrosis Factor Are Not Essential Parameters of CD4+ T-Cell Responses for Vaccine Control of
Tuberculosis. J Infect Dis 2015;212:495–504. doi:10.1093/infdis/jiv055.
[23] Orr MT, Beebe EA, Hudson T, Argilla D, Huang P-WD, Reese VA, et al. Mucosal delivery switches the
response to an adjuvanted tuberculosis vaccine from systemic TH1 to tissue-resident TH17 responses
without impacting the protective efficacy. Vaccine 2015;33:6570–8.
doi:10.1016/j.vaccine.2015.10.115.
[24] Sallin MA, Kauffman KD, Riou C, Bruyn ED, Foreman TW, Sakai S, et al. Host resistance to pulmonary
Mycobacterium tuberculosis infection requires CD153 expression. Nat Microbiol 2018:1.
doi:10.1038/s41564-018-0231-6.
[25] Vandewalle K, Festjens N, Plets E, Vuylsteke M, Saeys Y, Callewaert N. Characterization of genomewide ordered sequence-tagged Mycobacterium mutant libraries by Cartesian Pooling-Coordinate
Sequencing. Nat Commun 2015;6. doi:10.1038/ncomms8106.
[26] Master SS, Davis AS, Rampini SK, Keller C, Ehlers S, Springer B, et al. Mycobacterium tuberculosis
Prevents Inflammasome Activation. Cell Host Microbe 2008;3:224–32.
doi:10.1016/j.chom.2008.03.003.
[27] Titgemeyer F, Amon J, Parche S, Mahfoud M, Bail J, Schlicht M, et al. A Genomic View of Sugar
Transport in Mycobacterium smegmatis and Mycobacterium tuberculosis. J Bacteriol 2007;189:5903–
15. doi:10.1128/JB.00257-07.
[28] Sassetti CM, Boyd DH, Rubin EJ. Genes required for mycobacterial growth defined by high density
mutagenesis. Mol Microbiol 2003;48:77–84. doi:10.1046/j.1365-2958.2003.03425.x.
[29] Infante E, Aguilar LD, Gicquel B, Pando RH. Immunogenicity and protective efficacy of the
Mycobacterium tuberculosis fadD26 mutant. Clin Exp Immunol 2005;141:21–8. doi:10.1111/j.13652249.2005.02832.x.
[30] Hotter GS, Wards BJ, Mouat P, Besra GS, Gomes J, Singh M, et al. Transposon mutagenesis of Mb0100
at the ppe1-nrp locus in Mycobacterium bovis disrupts phthiocerol dimycocerosate (PDIM) and
glycosylphenol-PDIM biosynthesis, producing an avirulent strain with vaccine properties at least equal
to those of M. bovis BCG. J Bacteriol 2005;187:2267–77. doi:10.1128/JB.187.7.2267-2277.2005.
[31] Reddy TBK, Riley R, Wymore F, Montgomery P, DeCaprio D, Engels R, et al. TB database: an integrated
platform for tuberculosis research. Nucleic Acids Res 2009;37:D499–508. doi:10.1093/nar/gkn652.
[32] Saleh MT, Belisle JT. Secretion of an Acid Phosphatase (SapM) by Mycobacterium tuberculosis That Is
Similar to Eukaryotic Acid Phosphatases. J Bacteriol 2000;182:6850–3. doi:10.1128/JB.182.23.68506853.2000.
[33] Grundner C, Ng H-L, Alber T. Mycobacterium tuberculosis Protein Tyrosine Phosphatase PtpB
Structure Reveals a Diverged Fold and a Buried Active Site. Structure 2005;13:1625–34.
doi:10.1016/j.str.2005.07.017.
[34] Sellers RS, Clifford CB, Treuting PM, Brayton C. Immunological Variation Between Inbred Laboratory
Mouse Strains Points to Consider in Phenotyping Genetically Immunomodified Mice. Vet Pathol
Online 2012;49:32–43. doi:10.1177/0300985811429314.
27
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
[35] Doisne J-M, Soulard V, Bécourt C, Amniai L, Henrot P, Havenar-Daughton C, et al. Cutting Edge: Crucial
Role of IL-1 and IL-23 in the Innate IL-17 Response of Peripheral Lymph Node NK1.1− Invariant NKT
Cells to Bacteria. J Immunol 2011;186:662–6. doi:10.4049/jimmunol.1002725.
[36] Cooper AM, Mayer-Barber KD, Sher A. Role of innate cytokines in mycobacterial infection. Mucosal
Immunol 2011;4:252–60. doi:10.1038/mi.2011.13.
[37] Arbues A, Aguilo JI, Gonzalo-Asensio J, Marinova D, Uranga S, Puentes E, et al. Construction,
characterization and preclinical evaluation of MTBVAC, the first live-attenuated M. tuberculosis-based
vaccine to enter clinical trials. Vaccine 2013;31:4867–73. doi:10.1016/j.vaccine.2013.07.051.
[38] Chen JM, Islam ST, Ren H, Liu J. Differential productions of lipid virulence factors among BCG vaccine
strains and implications on BCG safety. Vaccine 2007;25:8114–22. doi:10.1016/j.vaccine.2007.09.041.
[39] Leung AS, Tran V, Wu Z, Yu X, Alexander DC, Gao GF, et al. Novel genome polymorphisms in BCG
vaccine strains and impact on efficacy. BMC Genomics 2008;9:413. doi:10.1186/1471-2164-9-413.
[40] Naka T, Maeda S, Niki M, Ohara N, Yamamoto S, Yano I, et al. Lipid phenotype of two distinct
subpopulations of Mycobacterium bovis Bacillus Calmette-Guerin Tokyo 172 substrain. J Biol Chem
2011;286:44153–61. doi:10.1074/jbc.M111.310037.
[41] Brosch R, Gordon SV, Garnier T, Eiglmeier K, Frigui W, Valenti P, et al. Genome plasticity of BCG and
impact on vaccine efficacy. Proc Natl Acad Sci U S A 2007;104:5596–601.
doi:10.1073/pnas.0700869104.
[42] Sander P, Clark S, Petrera A, Vilaplana C, Meuli M, Selchow P, et al. Deletion of zmp1 improves
Mycobacterium bovis BCG-mediated protection in a guinea pig model of tuberculosis. Vaccine
2015;33:1353–9. doi:10.1016/j.vaccine.2015.01.058.
[43] Results from Innovative Phase 2 Tuberculosis Vaccine Trial Offer Potential for New BCG Revaccination.
Aeras n.d. http://www.aeras.org//.
[44] Ernst JD. Mechanisms of M. tuberculosis Immune Evasion as Challenges to TB Vaccine Design. Cell
Host Microbe 2018;24:34–42. doi:10.1016/j.chom.2018.06.004.
[45] Brighenti S, Lerm M. How Mycobacterium tuberculosis Manipulates Innate and Adaptive Immunity –
New Views of an Old Topic. Underst Tuberc - Anal Orig Mycobacterium Tuberc Pathog 2012.
doi:10.5772/29651.
[46] Kalia V, Sarkar S, Ahmed R. CD8 T-cell memory differentiation during acute and chronic viral
infections. Adv Exp Med Biol 2010;684:79–95.
[47] Belisle JT, Sonnenberg MG. Isolation of genomic DNA from mycobacteria. Methods Mol Biol Clifton NJ
1998;101:31–44. doi:10.1385/0-89603-471-2:31.
[48] CLC bio. CLC bio Genomics Workbench - Manual n.d.
[49] D’haene B, Vandesompele J, Hellemans J. Accurate and objective copy number profiling using realtime quantitative PCR. Methods San Diego Calif 2010;50:262–70. doi:10.1016/j.ymeth.2009.12.007.
[50] Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor package for differential expression
analysis of digital gene expression data. Bioinformatics 2010;26:139–40.
doi:10.1093/bioinformatics/btp616.
[51] McCarthy DJ, Chen Y, Smyth GK. Differential expression analysis of multifactor RNA-Seq experiments
with respect to biological variation. Nucleic Acids Res 2012;40:4288–97. doi:10.1093/nar/gks042.
[52] Weischenfeldt J, Porse B. Bone Marrow-Derived Macrophages (BMM): Isolation and Applications.
Cold Spring Harb Protoc 2008;2008:pdb.prot5080. doi:10.1101/pdb.prot5080.
[53] Goude R, Roberts DM, Parish T. Electroporation of Mycobacteria. Mycobact. Protoc., Humana Press,
New York, NY; 2015, p. 117–30. doi:10.1007/978-1-4939-2450-9_7.
28
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
688
8. Figures
689
Figure 1| Whole genome resequencing of the WT BCG Pasteur 1721 strain and of the sapM::Tn BCG mutant.
690
Detailed representation of the read mappings at the sapM locus for both strains. The red square marks the
691
TA dinucleotide site where the transposon is inserted in the mutant strain.
692
Figure 2| (A) Volcano plot comparing the sapM::Tn BCG mutant and the WT BCG. Genes with a > 2-fold
693
change and a p-value < 0.001 are depicted in red. (B) Mapped (sense) reads at the SapM locus in the WT
694
BCG vs sapM::Tn BCG condition. Only one replicate of both WT BCG and sapM::Tn BCG samples is depicted.
695
BCG_3374c = upp, BCG_3375 = sapM. (C) RT-PCR analysis of the sapM locus. RNA was prepared of cultures
696
of biological triplicates of the sapM::Tn BCG disruption mutant, as well as of the WT BCG. An RT-PCR on the
697
cDNA using primer sets directed against the sapM gene (blue bar) and the directly up- and downstream genes
698
(upp and BCG_3376, orange and green bars, respectively). The data presented here are averages (± SD) of
699
the three biological replicates. For each mutant, the targets are individually normalized to the transcription
700
levels in the WT BCG strain (grey dotted line set at an average relative quantity of 1). The * indicates
701
significance (p < 0.05) by a two tailed t-test comparing the transcription levels of each target in the mutant
702
strain to the same target in the WT BCG. (D-F) SapM protein and activity analysis. The sapM::Tn BCG mutant
703
and WT BCG 1721 were grown in 7H9 medium and supernatant samples were collected on various time
704
points (D6 (Day 6) – D15). (D) ELISA using an anti-SapM polyclonal antibody. (E) SDS-PAGE and western
705
blotting of WT BCG and sapM::Tn BCG mutant supernatant samples collected on day 15 and grown in 7H9
706
medium without OADC and supplemented with glucose only. The blots were developed with the anti-SapM
707
antibody. The processed SapM protein (= removal of the N-terminal signal sequence) is ± 28 kDa in size and
708
is only present in the WT BCG strain (black arrow). (F) In vitro phosphatase assay using p-Nitrophenyl
709
phosphate (pNPP) as a substrate to check the activity of the SapM enzyme. The plotted data for both ELISA
710
and phosphatase assay are averaged over 2 technical replicates and corrected for the background signal
711
induced by sterile 7H9 medium supplemented with 10% OADC.
29
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
712
Figure 3 | SapM::Tn BCG safety study in SCID mice. Immunocompromised SCID mice were infected
713
intravenously with 3x106 versus 3x107 cfu of the WT BCG or SapM::Tn BCG mutant and survival was
714
monitored. (Log-rank test; WT BCG vs. sapM::Tn; P=not significant). The control groups represent animals
715
vaccinated with PBS.
716
Figure 4 | In vivo and in vitro replication analysis: sapM::Tn BCG versus WT BCG
717
(A) F1 mice (Balb/c x C57BL/6J) were vaccinated s.c. with the WT BCG or sapM::Tn BCG (2x106 cfu, 10
718
mice/group). At day 8, 14 and 28 post-infection, mice were sacrificed and the number of bacteria in the
719
draining LNs was determined by cfu plating (Mann-Whitney test; ** P<0.01). (B) BM-DMs were infected with
720
WT BCG or sapM::Tn BCG (MOI 10:1) for 4 hours. Cells were washed and bacterial uptake was determined
721
4hr post-infection. Bacterial replication was analyzed by cfu plating 24, 48 and 72hr, 6 days post-infection.
722
(C) F1 mice were vaccinated i.v. with the WT BCG or sapM::Tn BCG (2x106 cfu, 8 mice/group). At 24h and 3
723
weeks post-infection, mice were sacrificed and the number of bacteria in the spleens and lungs was
724
determined by cfu plating (Mann-Whitney test; ** P<0.01).
725
Figure 5 | RT-PCR analysis of the sapM locus and SapM protein and activity analysis in the
726
sapM::Tn:complementation mutant versus WT BCG
727
(A) RNA was prepared of cultures of biological triplicates of the sapM::Tn BCG disruption mutant (red), the
728
WT BCG (black) and the sapM::Tn:complementation BCG mutant (green). An RT-PCR on the cDNA used
729
primer sets directed against the sapM gene (sapM b primers - left panel) and the directly upstream gene upp
730
(upp2 primers - right panel). The data presented here are averages (± SD) of three technical replicates. For
731
each mutant, the targets are individually normalized to the transcription levels in the WT BCG strain (grey
732
dotted line set at an average relative quantity of 1). The * indicates significance (p < 0.05) by a t-test
733
comparing the transcription levels of each target in the mutant strain to the same target in the WT BCG. (B-
734
D) The sapM::Tn BCG mutant, WT BCG 1721, sapM::Tn:complementation BCG mutant and sapM::Tn:empty
735
vector were grown in 7H9 medium and supernatant samples were collected on various time points (D1 (Day
736
1) – D15). (B) ELISA using an anti-SapM polyclonal antibody. The plotted data for both ELISA and phosphatase
30
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
737
assay is averaged over 2 technical replicates and corrected for the background signal induced by sterile 7H9
738
medium supplemented with 10% OADC. (C) SDS-PAGE and western blotting of indicated supernatant samples
739
collected on day 10 and grown in 7H9 medium without OADC and supplemented with glucose only. The blots
740
were developed with the anti-SapM antibody. The processed SapM protein (= removal of the N-terminal
741
signal sequence) is ± 28 kDa in size and is only present in the WT BCG strain and sapM::Tn:complementation
742
mutant. (D) In vitro phosphatase assay using p-Nitrophenyl phosphate (pNPP) as a substrate to check the
743
activity of the SapM enzyme.
744
Figure 6| In vivo replication analysis: sapM::Tn BCG versus sapM::Tn:complementation BCG versus WT BCG
745
(A) C57BL/6J mice were vaccinated s.c. with the WT BCG, sapM::Tn BCG or sapM::Tn:complementation BCG
746
(2x106 cfu, 6 mice/group). At day 8, 14 and 28 post-infection, mice were sacrificed and the number of bacteria
747
in the draining LNs was determined by cfu plating (Mann-Whitney test; ** P<0.01). (B) C57BL/6J mice were
748
vaccinated i.v. with the WT BCG, sapM::Tn BCG or sapM::Tn:complementation BCG (2x106 cfu, 7-8
749
mice/group). At 24h and 3 weeks post-infection, mice were sacrificed and the number of bacteria in the
750
spleens and lungs was determined by cfu plating (Mann-Whitney test; ** P<0.01).
751
Figure 7 | iDC recruitment to lymphoid organs starts earlier when mice are vaccinated with the BCG
752
sapM::Tn BCG strain
753
C57BL/6J mice were immunized s.c. at the base of the tail (2x106 cfu, 5-10 mice/group (A-B); 2.105 cfu, 10
754
mice/group (C-D) ) with WT BCG or sapM::Tn BCG mutant. Seven and 14 days later, the mice were sacrificed,
755
and inguinal LNs, brachial LNs and spleens were isolated. Cells were prepared, labelled with different
756
antibodies staining DCs and analyzed by flow cytometry. Total cell numbers (A, C) and the kinetics of iDC
757
(CD103+MHCIIhi CD11b+ Ly6C+) recruitment (B, D) differ upon BCG sapM::Tn BCG and WT BCG vaccination.
758
(Mann-Whitney; *P<0.05, **P<0.01).
759
Figure 8 | Vaccination with sapM::Tn BCG induces reduced frequencies of IFNγ-producing CD4+ and CD8+ T
760
cells compared to WT BCG
31
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
761
C57BL/6J mice were immunized s.c. at the base of the tail with a high dose of WT BCG (black bars) or sapM::Tn
762
BCG mutant (red bars) (2x106 cfu, 5 mice/group). 14 days later, the mice were sacrificed, and inguinal LNs,
763
brachial LNs and spleens were isolated. Cells were prepared and the T cell response was analyzed by
764
intracellular cytokine staining followed by flow cytometry (A). Additionally, supernatants of stimulated
765
spleen cells and controls were harvested 48hr post-stimulation and IFNγ concentration was determined by
766
bioplex (B). (Mann-Whitney or 2way ANOVA; *P<0.05, **P<0.01).
767
Table 1 | Variants detected in WT BCG and the sapM::Tn BCG disruption mutant.
768
A probabilistic variant analysis was performed in CLC genome workbench on the reads mappings of both the
769
WT BCG and sapM::Tn BCG mutant strain. We observed only very few variations compared to the M. bovis
770
BCG Pasteur str. 1173P2 reference genome. (SNV/MNV = single/multiple nucleotide variants; Indel =
771
insertions or deletions). The Tn insertion was picked up using de novo assembly of the sapM::Tn reads.
32
Fig. 1
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
Fig. 2
A.
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
B.
C.
E.
D.
F.
Fig. 3
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
Fig. 4
A.
B.
C.
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
Fig. 5
50 kDa
37 kDa
25 kDa
20 kDa
SapM::Tn
WT
C.
SapM::Tn:empty vector
bioRxiv preprint doi: https://doi.org/10.1101/486993. this versionB.
posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
SapM::Tn:Compl
A.
D.
Fig. 6
A.
B.
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
Fig. 7
A.
C.
B.
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
D.
Fig. 8
A.
B.
bioRxiv preprint doi: https://doi.org/10.1101/486993. this version posted December 4, 2018. The copyright holder for this preprint (which was
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
Vaccination-site draining lymph node; 2 weeks post-vaccination
Spleen; 2 weeks post-vaccination
Note: in all figures:
Variant 6a
AA change
K43R
Synonymous
L85fs
A433InsA
G261Deletion
3,854,971
InDel
cut3
Insertion GR
3,685,759 Tn insertion SapM promoter
3,708,425
InDel
sugI
S13fs
Variant 6b
3,708,517
Variant 1
Variant 2
Variant 3
Variant 4
Variant 5
Position
813,096
1,344,671
2,764,157
3,197,940
3,833,491
Mutation
Type
SNV
MNV
SNV*
InDel
InDel**
SNV
Gene
rpsL
BCG_2507c
fadD26
BCG_3499c
sugI
L44M
Illumina resequencing
1173P2 WT
1721
(reference)
SapM
::Tn
1721 BCG WT
A
G
G
CG
GC
GC
T
A
A
A
A
CGC
CGC
1173P2 WT
1721
SapM ::Tn
A
GC
1721 BCG WT
G
GC
-
A
A
G
GC
-
TCG
A
TCG
tn
-
TCG
-
TCG
TCG
A
-
C
C
A
C
C
A
* Not verified by Sanger sequencing, because the mutation is synonymous
** Sequencing did not work, due to the presence of repeats in the region of the mutation
Table 1
Sanger analysis