Figure 2: Micro-neutralization tests using different OPXVs. Five sera
from MPXV-infected (M1-M5) and six sera from CPXV-infected persons
(C1-C6) plus five sera from IMVANEX-vaccinated individuals (V1-V5) were
analysed by using TCID50 assays with different viruses.
Since IFA and NT are both low-throughput methods to detect binding or
neutralising antibodies, they are not suited to screen larger sample
numbers, e.g., in an outbreak situation; therefore, for this purpose an
ELISA was updated that had previously been established in the lab
[11]. The ELISA uses VACV-infected HEp2 cells as the specific
antigen and non-infected HEp2-cells as the negative control. Hereby,
unspecific or background binding can be eliminated effectively by
subtracting non-specific HEp2 signals from VACV-specific signals. To
render the ELISA also useful for the detection of acute infections, IgM
analysis was newly established. Initial experiments for IgM detection
using serum dilutions in the casein blocking buffer (also used for serum
dilution prior to IgG detection) led to high unspecific binding. Hence,
different blocking buffers and commercially available diluents were
tested (supporting figure 1). As LowCross buffer was most efficient in
reducing high background signals while retaining specific binding of
IgM-positive sera, it was used as the serum diluent in the final assay
protocol.
To test the agreement between the established IFA and the modified or
newly established IgG and IgM detection by ELISA, the quantified ELISA
results for 28 sera were compared with the corresponding IFA titres
(Figure 3). For IgG detection, a high level of correlation was found
between IFA titres and ELISA results with higher titres, leading to
higher ELISA results. This was also true for IgM detection, although
differentiation between lower titres was not possible.
Figure 3: Comparison of IgG (A) and IgM (B) titres by IFA with results
obtained by ELISA. IFA titres were determined by titration against CPXV-
or VACV-infected HEp2 cells.
After demonstrating that IFA and ELISA, moreover, using less pathogenic
OPXVs like VACV as antigens, can be used to detect (cross-reactive)
antibodies against MPXV, the next aim was to further enhance the
method’s biosafety with a robust inactivation method for the serum
samples. A literature review resulted in two potential protocols using
either 0.3 %Triton/0.3 %Tween/1 %TNBP (protocol 1, p1) [23] or
0.5 %Triton/0.5 %Tween/1h60 °C (protocol 2, p2) [22]. For
comparison, standard heat inactivation was used for 30 min at 56 °C.
First the protocols were tested with VACV, demonstrating that both
protocols resulted in a depletion of infectious poxvirus particles
(Figure 4A). However, the protocol p2 (Figure 4A) led to a better
inactivation compared to protocol p1. Heat inactivation alone was least
efficient in inactivating infectious virus particles (Figure 4A, heat).
Since protocol p2 worked well the next aim was to reduce the heat
inactivation time and reduce the incubation temperature to allow
inactivation while still retaining the antibody-binding ability. For
this purpose, p2 was tested with classical heat inactivation conditions
(30 min 56 °C, protocol 3, p3) which led to a depletion comparable with
the p2 protocol (Figure 4, p2 and p3). Next, three replicates of VACV
inactivated with p3 were passaged three times in cell culture and
checked for viral DNA by real-time-PCR in the cells and the supernatant.
This resulted in verifying that in the 10-1 dilution
of the TCID50 there are few infectious particles left,
while in the 10-2 dilution no infectious particles
could be detected in all three replicates. Taken together, using the p3
protocol, a depletion in infectious virus particles of about 5 log-steps
could be shown. The inactivation potential of the p3 protocol was then
also verified with MPXV (Figure 3B).