Unique targeted testing of the urogenital microbiome has potential as a predictive test during IVF for implantation outcome.

Abstract

The aim of this pilot study was to develop a method characterising the urogenital microbiome as a predictive test in the IVF workup. Using unique custom qPCRs we tested for the presence of specific microbial species from vaginal samples and First Catch Urines from the male. The test panel included a range of potential urogenital pathogens, STIs, ‘favourable’ (Lactobacilli spp.) and ‘unfavourable’ bacteria (anaerobes) reported to influence implantation rates.
We tested couples attending Fertility Associates, Christchurch, New Zealand for their first round of IVF and found that some microorganisms affected implantation.
The qPCR result was interpreted qualitatively using the Z proportionality test. Samples from women at the time of Embryo Transfer who did not achieve implantation had significantly higher percent of samples that were positive for Prevotella bivia and Staphylococcus aureus compared to women who did achieve implantation.
The results provide evidence that most microorganisms chosen for testing had little functional effect on implantation rates. The addition of further microbial targets (yet to be determined) could be combined in this predictive test for vaginal preparedness on the day of Embryo Transfer. This methodology has a substantial advantage of being affordable and easily performed in any routine molecular laboratory.
This methodology is most suitable as a foundation on which to develop a timely test of microbiome profiling. Using the indicators detected to have a significant influence, these results can be extrapolated to a rapid antigen test for a woman to self-sample prior to Embryo Transfer as an indicator of likely implantation.

Full text

Page 1/19
Unique targeted testing of the urogenital microbiome has
potential as a predictive test during IVF for implantation
outcome.
Gloria Evelyn Evans (  gloria.evans@auckland.ac.nz )
University of Auckland Faculty of Medical and Health Sciences https://orcid.org/0000-0001-5804-4071
Vishakha Mahajan
Sarah Wakeman
Tania Slatter
Anna Ponnampalam
Trevor Anderson
Makhdoom Sarwar
John Evans
Research Article
Keywords: IVF, urogenital microbiome, predictive test, implantation rates
Posted Date: October 25th, 2022
DOI: https://doi.org/10.21203/rs.3.rs-2136685/v1
License:   This work is licensed under a Creative Commons Attribution 4.0 International License.  Read Full
License

Page 2/19
Abstract
The aim of this pilot study was to develop a method characterising the urogenital microbiome as a predictive test in
the IVF workup. Using unique custom qPCRs we tested for the presence of specic microbial species from vaginal
samples and First Catch Urines from the male. The test panel included a range of potential urogenital pathogens,
STIs, ‘favourable’ (
Lactobacilli
spp.) and ‘unfavourable’ bacteria (anaerobes) reported to inuence implantation
rates.
We tested couples attending Fertility Associates, Christchurch, New Zealand for their rst round of IVF and found
that some microorganisms affected implantation.
The qPCR result was interpreted qualitatively using the Z proportionality test. Samples from women at the time of
Embryo Transfer who did not achieve implantation had signicantly higher percent of samples that were positive
for
Prevotella bivia
and
Staphylococcus aureus
compared to women who did achieve implantation.
The results provide evidence that most microorganisms chosen for testing had little functional effect on
implantation rates. The addition of further microbial targets (yet to be determined) could be combined in this
predictive test for vaginal preparedness on the day of Embryo Transfer. This methodology has a substantial
advantage of being affordable and easily performed in any routine molecular laboratory.
This methodology is most suitable as a foundation on which to develop a timely test of microbiome proling. Using
the indicators detected to have a signicant inuence, these results can be extrapolated to a rapid antigen test for a
woman to self-sample prior to Embryo Transfer as an indicator of likely implantation.
Capsule
The urogenital microbiome of couples attending their rst IVF was compared to implantation rates. Unique custom
qPCR methodology detected signicantly different levels of microorganisms in women who did not achieve
implantation.
1 Introduction
Lack of implantation of an embryo into the endometrium in women undergoing IVF is far too common an
occurrence. Implantation rates worldwide are still disappointingly low [1] and few reliable predictive tests are
available.
Researching the urogenital microbiome using Next Generation Sequencing (NGS) of the 16S ribosomal RNA gene
has been reported [2]. The major ndings were that women whose vagina was dominated by a variety of
Lactobacilli species had the most favourable pregnancy outcome [2] and is still considered an indicator of vaginal
health [3] by maintaining an acidic pH to aid protection from the effect of anaerobes and
Escherichia coli
[4].
Further, women who harboured anaerobes notably
Gardnerella vaginalis
[5, 6] causing Bacterial/Anaerobic
vaginosis and potential pathogens
Enterococcus sp., Escherichia coli
and/or
Streptococcus sp.
also had a less
favourable outcome [2, 3]. Findings supported by others, in women [7–10] and in the male [11].
Because of the diculty with recruitment we report this work as a pilot study. This study investigated this concept
that a range of specic microbial species present in the urogenital tract of couples undergoing their rst round of

Page 3/19
IVF could affect the implantation rates of the transferred embryo.
This project determined to understand the biodiversity of the microbiome reported to inuence IVF outcome. Our
aim was to initiate the development of a rapid, affordable, predictive test of microbiome proling in the routine IVF
workup. Individual microbiome proling could have the potential to assist the couple in deciding whether to
continue IVF in that cycle.
Our testing panel included
Lactobacilli
species, anaerobes, potential urogenital pathogens and STIs. Additional
testing also included the most common STI, Human Papilloma Virus (HPV). So far, 18 high-risk and 12 low-risk
subtypes of genital HPV have been identied [12]. HPV has been implicated in inuencing IVF outcome in women
[13, 14] and in the male [15] although others have reported this association to be less clear [16]. Of the STIs
included for testing
Ureaplasma urealyticum, Ureaplasma parvum
and
Mycoplasma hominis
have been reported to
contribute to the condition of anaerobic vaginosis [4, 17] and been implicated in affecting reproduction outcomes
[18, 19]. Further,
Mycoplasma genitalium
has also been reported in couples experiencing failed IVF [20].
2 Methods
This study received the approval of the Southern Health and Disability Ethics Committee, New Zealand application
number 15/STH/65.
2.1 Sample population:
Samples were collected prospectively from a heterogenous group of 32 couples who attended Fertility Associates
clinic in Christchurch, New Zealand and satised the following criteria.
Inclusion criteria included: Couples aged 20–40 without any confounding health issues, who had not taken
antibiotics in the previous month, were non-smokers and undergoing their rst, fresh IVF cycle
Exclusion criteria included: Women on frozen embryo transfer cycles. Male partners with frozen semen.
2.2 Samples collected:
Women had two vaginal samples collected, the rst obtained by self-collection and the second collected by the
clinician [21]:
1. Baseline Sample A, Collected The Cycle Before The Ivf Cycle, In The Mid
Luteal Phase.
2. Sample B At Fresh Embryo Transfer (Et).
and
3. Baseline Sample C from the male partner, fresh semen or a First Catch Urine (FCU). A further semen sample was
collected at ET in MycoDuo media for Mycoplasmataceae culture.
Vaginal samples were collected repurposing swabs for molecular testing uing the BD ProbeTec Qx collection kit
441357, Cat. 22-370-171 (Fisher Scientic).

Page 4/19
Of the 32 couples who consented to participate in this project, 2 were excluded as they did not proceed on to IVF the
following cycle. Therefore n = 30 couples.
2.3 Techniques employed for the detection of microbial
organisms:
1. Nugent Gram stain scores were assessed for visible bacterial populations present in vaginal swabs A and B
[22].
2. qPCR for molecular presence or absence of microbial species in samples A, B and C [23].
3. Molecular detection for the presence or absence of
Human Papilloma Virus
(HPV) - with subtyping, using the
Euroimmun HPV typing array (Perkin Elmer) [12] in samples A, B C or semen in MycoDuo media (ET).
4. Culture of male partner’s semen sample (at ET) collected in MycoDuo medium (BioRad) [24] for the detection
of
Mycoplasma hominis, Ureaplasma urealyticum
and
(Ureaplasma parvum).
2.3.1 Nugent Gram stain scores obtained from vaginal swabs:
Gram staining was performed on all vaginal swabs. WBC, Lactobacilli, Gardnerella and anaerobes were noted [22].
Squamous epithelial cells were noted to conrm the sample had been well collected. Scores were allocated from 0
to 4 where 0 indicated no anaerobes present but a predominance of Lactobacilli, whereas a value of 4 indicated a
predominance of
Gardnerella vaginalis
± other anaerobes.
2.3.2 Molecular testing platform for microbial species:
Unique Qiagen, Custom Microbial DNA qPCR Arrays Cat No. 330161 CBAID00051 (Qiagen) [23] are a user dened
assay developed specically for this project. PCR detected the bacterial 16S rRNA gene. Probes were designed for
speciic targets that were user-dened microbial species. (Table1).

Page 5/19
Table 1
Microbial targets chosen for qPCR testing
Gene Symbol Target name for microbial species NCBI taxonomy ID
A.Prev Anaerococcus prevotii 33034
C.Trac Chlamydia trachomatis 813
E.Faecalis Enterococcus faecalis 1351
F.Magn Finegoldia magna 1260
G.Vagi Gardnerella vaginalis 2702
L.Cris Lactobacillus crispatus 47770
L.Gass Lactobacillus gasseri 1596
L.Iner Lactobacillus iners 147802
L.Jens Lactobacillus jensenii 109790
M.Geni Mycoplasma genitalium 2097
M.Homi Mycoplasma hominis 2098
N.Gono Neisseria gonorrhoeae 485
P.Bivi Prevotella bivia 28125
S.Aure Staphylococcus aureus 1280
S.Agal Streptococcus agalactiae 1311
T.Vagi Trichomonas vaginalis 5722
U.Parv Ureaplasma parvum 2130
U.Urea Ureaplasma urealyticum 1352
E.Faecium Enterococcus faecium 1352
E.Coli Escherichia coli 623
S.Pyog Streptococcus pyogenes 1314
GAPDH Hs GAPDH 
Pan1 Pan Bacteria 1 
PPC Positive PCR Control 
The molecular testing platform wase categorised into 5 groups:
1. anaerobes -
Anaerococcus prevotii, Finegoldia magna, Gardnerella vaginalis
(facultative anaerobe),
Prevotella
bivia.
2.
Lactobacilli
spp. -
Lactobacillus crispatus, Lactobacillus gasseri, Lactobacillus iners, Lactobacillus jensenii.
3. Potential urogenital pathogens -
Enterococcus faecalis, Enterococcus faecium, Escherichia coli,
Staphylococcus aureus, Streptococcus agalactiae, Streptococcus pyogenes
.

Page 6/19
4. STIs -
Chlamydia trachomatis, Mycoplasma genitalium, Mycoplasma hominis, Neisseria gonorrhoeae,
Ureaplasma parvum
and
Ureaplasma urealyticum
and the parasite
Trichomonas vaginalis. Human papilloma
virus
with subtyping was tested for on a separate platform.
5. Three PCR controls were included for each sample, PCR Positive Control, and probes for both pan bacteria and
GAPDH.
Microbial DNA extraction (mDNA).
Vaginal swabs, semen or FCU deposits containing mixed DNA, had microbial DNA (mDNA) extracted using the
QIAmp UCP Pathogen Minikit #50214 [25].
Briey, the deposit of each sample was pre-treated in pathogen lysis tubes with Buffer ATL (containing Reagent Dx)
and vortexed to assist lysis of microbial species. The supernatant was then added to Proteinase K and incubated at
56°C. Buffer APL2, and was then added and further incubated. The lysate was precipitated with ethanol and applied
to the QIAmp UCP mini spin column. The column was washed with two buffers (APW1 and APW2) and the
resultant puried microbial DNA eluted off the column with Buffer AVE.
Testing platform for molecular testing of microbial species.
The 96-well microtitre plates were produced in 24 x 4 samples format. Microbial qPCR mastermix Format A with
Rox a passive reference dye was used containing PCR primers that detected the bacterial 16S rRNA gene. The
amplied product was detected using target-specic uorescent hydrolysis probes present in designated wells.
qPCR set up.
A standardised amount of 125ng of mDNA was recommended and used per reaction. This constant input of DNA
allowed for comparison of results.
A total volume of 25µL reaction mix was added per well.
qPCR was performed on the Quant Studio 6 Flex, Applied Biosystems platform (Life Technologies).
Thermocycling conditions: activation 10min 95°C 1 cycle, 2-step cycling of 45 cycles – denaturation 15 sec 95°C,
annealing and extension 2 min 60°C.
The threshold cycle (Ct) was calculated for each well for data analysis.
2.3.3 Molecular testing for HPV:
The EUROarray HPV testing platform detected 30 genital HPV types Table2, (EUROIMMUN, Perkin Elmer) [12].

Page 7/19
Table 2
HPV types tested using
the EUROarray testing
platform
HPV
subtypes
detected
18 high
risk HPV
12 low
risk HPV
16 6
18 11
26 40
31 42
33 43
35 44
39 54
45 61
51 70
52 72
53 81
56 89
58 
59 
66 
68 
73 
82 
Viral DNA extraction.
Viral DNA was extracted using the QIAmp DNA Mini Protocol #51304 (Qiagen).
Briey, vaginal swabs were treated with Proteinase K and Buffer ATL and incubated at 56°C for 16 hours to assist
lysis. The lysate was precipitated with ethanol and applied to the QIAmp mini spin column. The column was
washed with two buffers (AW1 and AW2) and the resultant puried viral DNA eluted off the column with Buffer AE.
Semen and FCU deposits were treated with a different extraction using the High Pure PCR Template Preparation Kit
# 11796828001 (Roche) [26].

Page 8/19
Briey, the sample deposit was treated with Binding Buffer and Proteinase K and incubated at 70°C. The lysate was
precipitated with isopropanol and applied to a High Pure Filter tube. The column was washed with Inhibition
Removal Buffer then Wash Buffer. The resultant puried viral DNA was eluted off the column using Elution Buffer.
qPCR set up.
As per instructions of product number MN 2540 − 2005 (EUROIMMUN, Perkin Elmer).The Microarray platform
detected oncogenes E6/E7 [12] using subtype specic primers and probes.
2.3.4 Culture for the detection of
Mycoplasma hominis
and
Ureaplasma urealyticum (
and
Ureaplasma parvum)
:
Samples were cultured according to the manufacturer’s instructions [24].
Briey, 0.2mL of semen or FCU was placed in Mycoplasma Duo culture medium. This culture uid was inoculated
directly into one well of a microplate to detect the presence of Mycoplasmas with titres of ≤ 103 CCU/mL, which
were considered as being present in a commensal role. The second well contained further diluted sample to detect
growth of Mycoplasmas at ≥ 104 CCU/mL, considered indicative of a pathogenic level of Mycoplasmas [24]. The
microplate was then incubated at 37°C for up to 48 hours. The presence of
Mycoplasma hominis
was noted when
the test wells containing arginine were hydrolysed and the phenol red indicator was changed to red, similarly,
Ureaplasma urealyticum/Ureaplasma parvum
hydrolysed urea within the well also changed the indicator red.
It should be noted that according to the manufacturer’s instructions the Mycoplasma Duo kit can also detect
Ureaplasma parvum
in the
U. urealyticum
well.
U. parvum
was originally a Biovar of
U. urealyticum
before it was
proposed to be renamed as a distinct species based on phylogenetic analysis [24, 27].
2.4 Data analysis:
2.4.1 Quantitative level of microbial species detected.
The presence of a particular microbial species was determined using the crossing point (Ct) where a positive signal
had crossed the baseline. Samples that crossed the baseline at 20–22 cycles were awarded a value of 10,
compared to a sample which required 38–40 cycles for detection were awarded a value of 1.
2.4.2 Qualitative analysis of the percent of samples that were
positive.
The data was converted to being positive (+) or negative (-) for each microorganism. The percent of samples with a
positive signal was estimated.
From this data, a one tailed Z proportionality test was used to detect the differences between the study groups
where a
p
values of < 0.05 was considered signicant. Samples collected at the same time were compared –
Samples A+ (implantation) were compared to Sample A- (no implantation) etc..
2.4.3 Data analysis of testing for HPV
Analysis and interpretation was fully automated using EUROArrayScan software [12].
3 Results

Page 9/19
Thirty couples were retrospectively categorised into two groups, those who achieved implantation (implantation
only, implantation and pregnancy only or a live birth) n = 15, and those who did not achieve implantation n = 15.
3.1 Nugent Gram stain scores obtained from vaginal swabs.
Gram staining indicatied there was minimal difference in Nugent scores from vaginal swabs taken at baseline (A)
and ET (B), from women who achieved implantation and those who did not (Table 3). In fact, most Nugent scores
tended to be predominantly anaerobic from both groups of women.

Women who
achieved
implantation
Nugent
score mean
Women who
achieved
implantation
with a
predominance
of Lactobacilli
detected by
qPCR
Women who
achieved
implantation
with a
predominance
of anaerobes
detected by
qPCR
Women who
did not
achieve
implantation
Nugent
score mean
Women who
did not
achieve
implantation
with a
predominance
of Lactobacilli
detected by
qPCR
Women who
did not
achieve
implantation
with a
predominance
of anaerobes
detected by
qPCR
Table 3
Nugent scores from vaginal swabs from women sampled at the baseline cycle (Sample A) and at embryo transfer
(Sample B) for those who achieved implantation and those who did not. The percent of samples that were positive
for
Lactobacilli
sp. and anaerobes detected by qPCR is added for comparison.
Sample
A
baseline
vaginal
swab
2.6 33% 40% 2.7 33% 27%
Sample
B
embryo
transfer
vaginal
swab
2.9 7% 47% 3.1 13% 40%
In women who had a Nugent score indicating anaerobes, none of the four urogenital anaerobes tested for were
found in 4 of 15 women who achieved implantation and 2 of 15 women who did not. Suggesting, that the
anaerobes seen microscopically were not selected to be included in the qPCR testing panel.
Three further distinctive anaerobes
Mobiluncus sp.
,
Peptosptreptococcus sp.
and
Atopobium vaginae
implicated in
anaerobic vaginosis were not included in the qPCR panel as they can be easily identied by their specic
morphology, were not seen.
3.2 Molecular testing for microbial species.
qPCR for microbial species detected organisms with increased sensitivity compared to traditional microscopy.
Table 3 demonstrates a comparison of molecular detection to a Nugent score.

Page 10/19
Table 3 also indicates that for Sample A only a third of women in both groups (implantation or no implantation)
had a predominant population of
Lactobacilli
sp. in baseline samples, dropping to just one or two samples for
Sample B from both groups of women.
It was noted that the STIs
Chlamydia trachomatis
and
Neisseria gonorrhoeae
and potential urogenital pathogens
Enterococcus faecium
and
Streptococcus pyogenes
were not detected in any samples by qPCR. Therefore, these
bacteria were excluded from the data analysis.
The most commonly detected bacterium in both groups was
Lactobacilli crispatus
. It was more common for a
woman to have the same mircoorganism detected or not detected. Thus,
levels
of microbial species detected
altered, but not their
presence
.
3.2.1 Quantitative level of microbial species detected.
Samples from the two groups of couples were quantitatively compared using the students
t
test.
Individual samples.
1. In Samples A and B, there was no signicant difference between the two groups in the means for each microbial
species.
2. Sample C, only
Lactobacilli crispatus
was signicantly different in the male partners of women who had not
achieved implantation
p
= 0.04.
Dynamic characteristics of the microbiome.
3. In women who achieved implantation levels of microbial species - Sample A vs B were compared to determine
the variation in microbial ora. Anaerobe
Finegoldia magna p
= 0.02 had signicantly different levels higher in
Sample A.
4. In women who did not achieve implantation - Samples A v B were compared. Anaerobes
Anaerococcus prevotii
p
= 0.05 and
Finegoldia magna
p
= 2x10− 4 had signicantly different levels higher in Sample A.
5. In couples who achieved implantation - Sample A vs C were compared. Anaerobes
Anaerococcus prevotii p
=
0.03,
Finegoldia magna p
= 7x10− 4 and
Prevotella bivia
p = 6 x10− 4, and
Lactobacilli jensenii p
= 8x10− 5 and
Lactobacilli gasserei p
= 0.03 had signicantly different levels which were higher in Sample A.
6. In couples who did not achieve implantation, - Samples A v C were compared. Anaerobes
Anaerococcus prevotii
p
= 9x10− 5 and
Finegoldia magna p
= 1x10− 4, and
Lactobacilli gasserei p
= 0.001 and
Lactobacilli iners p
= 0.03 had
signicantly different levels which were higher in Sample A.
The results in paragraphs iii) and vi) led us to believe that a woman self-sampling (Sample A) possibly had a more
effective sampling style for obtaining vaginal epithelial cells to which microbial species often adhere. Despite
levels
of microbial species detected being increased, this did not often correlate with the success or failure of
implantation. Using quantitative levels revealed little useful information.

Page 11/19
3.2.2 Qualitative analysis of the percent of samples that were
positive.
Thus, alternative analysis was required. Samples from the two groups of couples were qualitatively compared
using the Z test.
i. The percent of positive signals were determined in women from both groups with Samples A v B (Fig. 1). A
reduction in anaerobic numbers in Sample B in both groups of women was noted whereas for
Lactobacilli sp
,
potential pathogens and STIs the levels detected remained stable demonstrating stability of some urogenital ora
during two luteal phases of the cycle [28]. Of the anaerobes tested, the percent of samples that were positive for
Anaerococcus prevotii, Finegoldia magna
and
Prevotella bivia
were more unstable between samples. Substantially
more bacteria were detected in the baseline sample A than in Sample B at ET. Interestingly,
Gardnerella vaginalis
levels tended to remain stable, possibly indicating that this bacterium did not inuence the success or failure of
implantation (Fig. 1). Further,
Gardnerella vaginalis
was the least commonly detected anaerobe. Variable signals for
anaerobes only, indicated that their presence was not an artefact.
ii. The percent of samples that were positive for all 3 sample sets were then compared and presented within
separate groups of microbial species – anaerobes, Lactobacillaceae, potential urogenital pathogens and
Mycoplasmataceae (including
Trichomonas vaginalis
) (Fig. 2A-D). Figure 2 indicates the high number of samples
that were positive for
Lactobacilli crispatus
and the moderate number of samples positive for
Lactobacilli iners
in
both groups indicating that these bacteria do not inuence success or failure of implantation.
iii. The percent of samples that were positive were then observed within each sample set where Sample A(+)
implantation was compared to Sample A(-) no implantation and so on (i.e. B + vs B- ; C + vs C-). Table 4
demonstrates where the Z test detected signicant differences between the study groups.
Table 4
The Z score
and
p
values
of microbial
species
present in
individual
samples (A,
B and C) of
fteen
couples who
did achieve
implantation
(+) and
fteen
couples who
did not (-)

Page 12/19
Comparison of Sample B + and B- produced a
p
value that was signicant for
Prevotella bivia
and
Staphylococcus
aureus.
The percent of samples that were positive were signicantly higher in women who did not achieve
implantation. Sample C produced a signicant
p
value for
Lactobacillus crispatus
where levels were higher for male
partners of women who did not achieve implantation.
3.3. Molecular testing for viral species on a separate platform
Human papilloma virus with subtyping.
HPV was detected in 5 women and 1 male of the total 30 couples, although there was an absence of HPV in their
partners. In the group who achieved implantation - one woman had a reported low-risk subtype, another woman had
3 subtypes detected of which one was high risk, a male partner also had a high risk subtype of HPV detected. Of the
two women with HPV detected who did not achieve implantation, one had a high risk subtype and the other a low
risk subtype detected. We were unable to detect any HPV DNA in one woman who achieved a live birth.
3.4. Culture for the detection of
Mycoplasma hominis,
Ureaplasma urealyticum (and Ureaplasma parvum).
The potential inuence of the male partner was further examined whereby a semen sample from the male partner
at ET was cultured for Mycoplasmas.
For women who achieved implantation, their partners’ samples were positive by culture for
Ureaplasma urealyticum
in 4 cases, none were conrmed by qPCR. For women who did not achieve implantation their partners displaying a
positive result in 4 cases, 3 were conrmed by qPCR. qPCR detected Mycoplasma species with increased sensitivity
in 13 couples who achieved implantation and in 15 couples who did not. Dual infections were commonly detected
by qPCR.
4 Discussion

Page 13/19
Although this pilot study where the numbers recruited are limited, this is in the common range of other IVF studies
[28–32]. We are investigating ‘if a panel of individual microorganisms is able to predict implantation success
during IVF.’
The aim of this study was to determine a foundation for an accessible, predictive test of the urogenital microbiome
proling to t into the routine IVF workup. We used a collection method easily accomplished by both patient and
clinician.
We found that using the traditional methodology of Nugent scores was of little assistance as a diagnostic tool to
indicate implantation outcome. Most vaginal samples from either group (i.e. implantation or no implantation)
exhibited a predominance of anaerobes, only a few samples were dominated by
Lactobacilli sp.
This technique is
still used and currently reported [4]. The lack of value of the Nugent scoring in this context motivated us to use the
qPCR technique.
The procedure for Microbial DNA qPCR Assay Kits is simple, and can be performed in any molecular laboratory with
a real-time PCR instrument. Thus, this approach is easily achievable, within the range of an IVF budget and
providing a timely result that can be clinically applicable.
First, we estimated the quantitative
levels
of microorganisms present and secondly, the qualitative presence or
absence of a microbial species.
Quantitative levels of bacterial detection in Samples A + vs A- and Samples B + vs B-. Only Samples C + vs C- noted
one
Lactobcailli sp.
that was signicantly higher in males of women who did not achieve implantation.
Interestingly, this nding is not supported by the observation of their female partners.
Quantitative levels of bacterial detection of Sample A vs Sample B. Signicant differences of some anaerobes and
some
Lactobacilli
sp. in both groups were found. It was noted that in individual women the level of microorganisms
in Sample A often varied from that of Sample B supporting the suggestion that a dynamic environment is involved
[21]. Of particular note is our observation that anaerobes were the main driver of bacterial dynamics. This
observation questions why different bacteria behave in a different manner and how the dynamics of the
microbiome [19–21] are regulated.
Further, potential pathogens,
Staphylococcus aureus
and
Streptococcus agalactiae
and STIs
Mycoplasma hominis
tended to be commonly detected in both groups (Fig. 1). Possibly suggesting that a degree of microbial diversity is
required to maintain basic physiological function of the urogenital tract.
Further, the quantitative
level
of microbial pathogens detected was very reliant on sample collection. Standardised
sampling in accessing vaginal epithelial cells is not applicable. Frequently when Sample B was collected by the
clinician, more mDNA had to be loaded to reach the same concentration of 125ng for testing. This factor
contributed to us simplifying the analysis of data to a qualitative analysis.
Qualitative analysis of qPCR results. The ndings were usually similar when comparing the microbial presence or
absence between couples who achieved implantation and those who did not. Samples generally had the same
microbial species detected either by both samples being positive or negative, with minimal variation, only a few
samples differed. These results indicated there was not rapid or frequent alteration of the qualitative nature of the
microbiome in these couples.

Page 14/19
The percent of samples that were positive suggests that the mere presence of a microorganism had an effect on
embryo implantation. A number of individuals who achieved implantation nurtured a level of anaerobes in the
urogenital tract commonly reported to be associated with non-implantation. However, the Z test produced two
signicant
p
values in Sample B, in women who did not achieve implantation. Increased levels of anaerobe
Prevotella bivia
, as conrmed by others [2, 7–10] and
Staphylococcus aureus
were detected. The lack of signicant
detection of other selected anaerobes such as
Gardnerella vaginalis
does not support these ndings. Possibly,
some anaerobes chosen to be included in the testing panel may not have been as active as others. Pro-
inammatory characteristics present in anaerobic/bacterial vaginosis [17, 33] as well as potential urogenital
pathogens can assist implantation [34]. Our nding indicates that while a proportion of ‘healthy’ microbial species
may be important, rates of implantation may be dependent on the biodiversity of microbial species. The implication
of such a conclusion is that the interactions and co-effects of one microorganism on another must be delineated in
the context of the ecient application of IVF. This unique, custom array testing panel allows the option to choose
new combinations and further microbial species.
The other signicant nding of
Staphylococcus aureus
also in Sample B of women who did not achieve
implantation could be explained by sampling technique. However, this trend was not noted in Sample B from
women who did achieve implantation.
Staphylococcus aureus
is considered as a skin contaminant. There were no
signicant differences detected for microbial species in Sample A. Sample C also had a signicant
p
value detected
for one
Lactobacilli sp.
in males of couples who did not achieve implantation.
The lack of HPV detection may be explained where negative partners may have resolved their HPV infection either
by vaccination or development of their own immunity [35]. The incidence of HPV detected was too low to ascribe
any inuence to implantation outcome and is not a robust predictor in our sample population.
The vaginal cavity has the capability to uctuate creating a dynamic urogenital microbiome environment [36–38]
as observed in this work, and is considered to be a likely inuence in the success of implantation rates [39, 40].
Further, an increase in microbial populations detected in women only, had no effect on implantation outcome. This
nding was possibly inuenced by the female urogenital tract being more hospitable to microbial species
compared to the male who may or may not have harboured a different or less hospitable environment to harbour
microbial species [41].
Further, the contribution of the microbiology of the male partner was also considered. Male Reproductive Proteins
(MRPs) can also have broad implications for successful reproduction. MRPs have the potential to inuence the
composition of the vaginal microbiome and thus the success of implantation [42]. But in this study, a wide variation
in the male urogenital microbiome often did not correspond with the women’s microbiome.
Conclusion
The results of this study challenges a concept of current thinking and is at the interface of research and clinical
application.
The unique methodology of this pilot project is most suitable as a foundation on which to develop an affordable,
timely test of microbiome proling in the routine IVF workup. Using the two indicators that were detected to have a
signicant inuence, these results can be extrapolated to a rapid antigen test for a woman to self-sample prior to

Page 15/19
ET as an indicator of likely implantation. The addition of further microbial targets (yet to be determined) can also
be combined in this predictive test for vaginal preparedness on the day of ET.
Declarations
Disclosure statement and Declaration for all authors
All authors state that they have no conicting interest to declare.
Article Type
Reproductive Biology.
Funding statement
This research was funded by grant 8.1.21 from the Maurice and Phyllis Paykel Trust, New Zealand.
Data availability statement
Not applicable.
Attestation statements
Data regarding any of the subjects in the study has not been previously published.
Data will be made available to the editors of the journal for review or query upon request.
Conict of Interest
None. Form attached.
Acknowledgement. The authors wish to thank the wonderful contribution of couples who participated in this project
as they underwent their rst round of IVF.
Thanks also to Mona Schousboe, Canterbury Health Laboratories, Christchurch, New Zealand who supported the
concept of this research, the staff of Fertility Associates, Christchurch, New Zealand who supported this project,
Aubrey Brand, Bio-strategy, New Zealand and David Reynolds, Qiagen, Australia who assisted with developing the
testing platform.
References
1. Gleicher, N., V.A. Kushnir, and D.H. Barad,
Worldwide decline of IVF birth rates and its probable causes.
Hum
Reprod Open, 2019. 2019(3): p. hoz017.
2. Hyman, R.W., et al.,
The dynamics of the vaginal microbiome during infertility therapy with in vitro fertilization-
embryo transfer.
J Assist Reprod Genet, 2012. 29(2): p. 105-15.
3. Baker, J.M., D.M. Chase, and M.M. Herbst-Kralovetz,
Uterine Microbiota: Residents, Tourists, or Invaders?
Front
Immunol, 2018. 9: p. 208.
4. Lin, Y.P., et al.,
Vaginal pH Value for Clinical Diagnosis and Treatment of Common Vaginitis.
Diagnostics
(Basel), 2021. 11(11).

Page 16/19
5. Swidsinski, A., et al.,
Presence of a polymicrobial endometrial biolm in patients with bacterial vaginosis.
PLoS
One, 2013. 8(1): p. e53997.
. Danielsson, D., P.K. Teigen, and H. Moi,
The genital econiche: focus on microbiota and bacterial vaginosis.
Ann
N Y Acad Sci, 2011. 1230: p. 48-58.
7. Koedooder, R., et al.,
The vaginal microbiome as a predictor for outcome of in vitro fertilization with or without
intracytoplasmic sperm injection: a prospective study.
Hum Reprod, 2019. 34(6): p. 1042-1054.
. Haahr, T., et al.,
Abnormal vaginal microbiota may be associated with poor reproductive outcomes: a
prospective study in IVF patients.
Hum Reprod, 2016. 31(4): p. 795-803.
9. Saraf, V.S., et al.,
Vaginal microbiome: normalcy vs dysbiosis.
Arch Microbiol, 2021. 203(7): p. 3793-3802.
10. Bernabeu, A., et al.,
Effect of the vaginal microbiome on the pregnancy rate in women receiving assisted
reproductive treatment.
J Assist Reprod Genet, 2019. 36(10): p. 2111-2119.
11. Schuppe, H.C., et al.,
Urogenital Infection as a Risk Factor for Male Infertility.
Dtsch Arztebl Int, 2017. 114(19): p.
339-346.
12. Euroimmun,
Chapter Molecular Infection Diagnostics. Section Sexually Transmitted Infections. Product
Catalgue
. 2021.
13. Pereira, N., et al.,
Human Papillomavirus Infection, Infertility, and Assisted Reproductive Outcomes.
J Pathog,
2015. 2015: p. 578423.
14. Yuan, S., et al.,
Human papillomavirus infection and female infertility: a systematic review and meta-analysis.
Reprod Biomed Online, 2020. 40(2): p. 229-237.
15. Garolla, A., et al.,
Spontaneous fertility and in vitro fertilization outcome: new evidence of human
papillomavirus sperm infection.
Fertil Steril, 2016. 105(1): p. 65-72.e1.
1. Schillaci, R., et al.,
Detection of oncogenic human papillomavirus genotypes on spermatozoa from male
partners of infertile couples.
Fertil Steril, 2013. 100(5): p. 1236-40.
17. Onderdonk, A.B., M.L. Delaney, and R.N. Fichorova,
The Human Microbiome during Bacterial Vaginosis.
Clin
Microbiol Rev, 2016. 29(2): p. 223-38.
1. Evans, G.E., et al.,
Evaluation of the Mycoplasma Duo kit for the detection of Mycoplasma hominis and
Ureaplasma urealyticum from urogenital and placental specimens.
Br J Biomed Sci, 2007. 64(2): p. 66-9.
19. Moragianni, D., et al.,
Genital tract infection and associated factors affect the reproductive outcome in fertile
females and females undergoing in vitro fertilization.
Biomed Rep, 2019. 10(4): p. 231-237.
20. Mohseni Moghadam, N., et al.,
Isolation and molecular identication of mycoplasma genitalium from the
secretion of genital tract in infertile male and female.
Iran J Reprod Med, 2014. 12(9): p. 601-8.
21. Forney, L.J., et al.,
Comparison of self-collected and physician-collected vaginal swabs for microbiome
analysis.
J Clin Microbiol, 2010. 48(5): p. 1741-8.
22. Nugent, R.P., M.A. Krohn, and S.L. Hillier,
Reliability of diagnosing bacterial vaginosis is improved by a
standardized method of gram stain interpretation.
J Clin Microbiol, 1991. 29(2): p. 297-301.
23. Qiagen,
Microbial DNA qPCR Handbook
. 2015.
24. Bio-Rad,
Mycoplasma Duo. Identication and Differential Titration of Genital Mycoplasma. 62739, 62740.
2010.
25. Qiagen,
QIAmp® UCP Pathogen Mini Handbook
. 2014.
2. Roche,
High Pure PCR Template Preparation Kit #117968280012008.

Page 17/19
27. Fanrong K, J.G., Zhenfang M, Gordon S, Wang B, Gilbert GL. ,
Phylogenetic analysis of Ureaplasma urealyticum
– support for the establishment of a new species, Ureaplasma parvum. .
International Journal of Systematic
and Evolutionary Microbiology, 1999. 49(4): p. 1879-1889.
2. Moreno, I., et al.,
Evidence that the endometrial microbiota has an effect on implantation success or failure.
Am
J Obstet Gynecol, 2016. 215(6): p. 684-703.
29. Carosso, A., et al.,
Controlled ovarian stimulation and progesterone supplementation affect vaginal and
endometrial microbiota in IVF cycles: a pilot study.
J Assist Reprod Genet, 2020. 37(9): p. 2315-2326.
30. Kitaya, K., et al.,
Characterization of Microbiota in Endometrial Fluid and Vaginal Secretions in Infertile Women
with Repeated Implantation Failure.
Mediators Inamm, 2019. 2019: p. 4893437.
31. Diaz-Martínez, M.D.C., et al.,
Impact of the Vaginal and Endometrial Microbiome Pattern on Assisted
Reproduction Outcomes.
J Clin Med, 2021. 10(18).
32. Riganelli, L., et al.,
Structural Variations of Vaginal and Endometrial Microbiota: Hints on Female Infertility.
Front Cell Infect Microbiol, 2020. 10: p. 350.
33. Boomsma, C.M., et al.,
Is bacterial vaginosis associated with a pro-inammatory cytokine prole in endometrial
secretions of women undergoing IVF?
Reprod Biomed Online, 2010. 21(1): p. 133-41.
34. Simon, C.,
Introduction: Do microbes in the female reproductive function matter?
Fertil Steril, 2018. 110(3): p.
325-326.
35. Innes, C.R., et al.,
Changes in human papillomavirus genotypes associated with cervical intraepithelial
neoplasia grade 2 lesions in a cohort of young women (2013-2016).
Papillomavirus Res, 2018. 6: p. 77-82.
3. Hickey, R.J., et al.,
Understanding vaginal microbiome complexity from an ecological perspective.
Transl Res,
2012. 160(4): p. 267-82.
37. Romero, R., et al.,
The composition and stability of the vaginal microbiota of normal pregnant women is
different from that of non-pregnant women.
Microbiome, 2014. 2(1): p. 4.
3. Srinivasan, S., et al.,
Temporal variability of human vaginal bacteria and relationship with bacterial vaginosis.
PLoS One, 2010. 5(4): p. e10197.
39. Ravel, J., et al.,
Vaginal microbiome of reproductive-age women.
Proc Natl Acad Sci U S A, 2011. 108 Suppl
1(Suppl 1): p. 4680-7.
40. Singer, M., et al.,
The relation of the vaginal microbiota to early pregnancy development during in vitro
fertilization treatment-A meta-analysis.
J Gynecol Obstet Hum Reprod, 2019. 48(4): p. 223-229.
41. Liu, C.M., et al.,
Male circumcision signicantly reduces prevalence and load of genital anaerobic bacteria.
mBio, 2013. 4(2): p. e00076.
42. Ness, R.B. and D.A. Grainger,
Male reproductive proteins and reproductive outcomes.
Am J Obstet Gynecol,
2008. 198(6): p. 620.e1-4.
Figures

Page 18/19
Figure 1
The percent of samples that were positive for each bacteria in women for Samples A and B

Page 19/19
Figure 2
A-D Percent of samples that were positive for couples who achieved implantation (A+, B+ and C+) and those who
did not for (A-, B-, C-)