Lin 1

Development of a qPCR Assay for Quantification of ​Saccharibacteria

By
John Lin

Boston Latin School,

78 Avenue Louis Pasteur
Boston, MA 02115

March 2019
 Lin 2

Acknowledgments

I would like to thank Dr. Andrew Collins and Dr. Pallavi Murugkar for introducing the basic lab
procedures that were critical to this project, including polymerase chain reaction (PCR), gel
electrophoresis, plasmid design, etc and for their expertise in the troubleshooting process. I am
also grateful for the members of the Dewhirst Lab and the Forsyth Institute for providing the lab
space and equipment for the project, funded under NIH grant ​R01DE024468-04. Finally, I would
like to thank Dr. Sowmya Balasubramanian, Dr. Megan Pugach-Gordon, Dr. Martin Taubman
and Anna Rifkin for additional support and mentorship.
 Lin 3

Table of Contents

Abstract 4

Introduction 5

Objectives 6

Criteria 6

Assay Development and Methods 7

Primer Design 7
Primer Optimization 7
Plasmid 7
qPCR 7
Application 7

Results 8
Primer Design and Optimization 8
Identification of Specific, Conserved Regions 8
Verification of Primer Specificity to Saccharibacteria Phylum 8
Primer Optimization 9
qPCR 10
Standard Curve and Efficiency 10
Primer Robustness Verification 11
Application 11

Discussion 12

References 13

Appendices 14
A. Plasmid Sequencing Data 14
B. qPCR 14
C. Growth Curve Data 15
 Lin 4

Abstract
Saccharibacteria (​ TM7) is an extremely small, uncultivable microbe that lives a unique parasitic
lifestyle on various host bacteria, including ​P. propionicum​. ​Saccharibacteria ​was recently found
in great abundance in plaque samples of patients with periodontal disease. There is​ ​limited
knowledge about the behavior of this intriguing microbe because of its evasiveness from
traditional, bacterial quantification methods due to its minuscule size and inability to be cultured
independently.

This study aims to develop a unique method to quantify this bacteria. Primers were designed and
optimized to target conserved regions, specific to the ​Saccharibacteria​ genome. A plasmid was
constructed using amplified regions of ​Saccharibacteria​ DNA and transformed in ​E. coli​. The
plasmid was extracted and diluted to a known number of plasmid copies. Serial dilutions were
performed and run on a qPCR. A calibration curve was developed based on the qPCR data.

In the end, a highly accurate assay was developed with 91.38% efficiency for quantifying
Saccharibacteria.​ These findings open the door to various applications. Clinically, the assay can
be used as a tool to diagnose and assess the risk for periodontitis based on the number of
Saccharibacteria​ present. In the research setting, the assay can be used to further examine the
behaviors of ​Saccharibacteria,​ its interactions with its host, and its pathogenicity.
 Lin 5

Introduction
The oral microbiome is a rich and diverse biosystem, containing over 700 species of bacteria.
Understanding this ecosystem, its inhabitants and their interactions is critical in developing more
effective treatments for oral diseases, including gingivitis and periodontal disease. While many
oral bacteria can be cultivated and grown ​in vitro,​ certain phyla display a recalcitrant nature to
independent cultivation. Among these are the ​Saccharibacteria​ phylum (also known as the TM7
candidate phylum). While ​Saccharibacteria​ has been found in natural habitats including
seawater, soil and human gut, it was identified for the first time to be part of the human
microbiome.

Saccharibacteria​ exists on as a 200nm, parasitic bacteria that lives an epibiotic lifestyle,

meaning that it lives on top of various hosts, including ​Actinomyces odontolyticus,​
Propionibacterium propionicum,​ etc. Due to its reliance on specific, unidentified nutrients from
its host, ​Saccharibacteria​ remains unable to be cultivated independently; despite this, a coculture
of the parasite and its host can be isolated and grown in nutrient-rich media, including Trypticase
Soy Broth with Yeast extract (TSBY) and SHI media.

Recently classified as a red complex bacteria, meaning that it is present in severe stages of
periodontitis, ​Saccharibacteria​ has demonstrated potential pathogenicity. An analysis of the 16S
DNA profile of an oral microbiome suggests that ​Saccharibacteria​ DNA is found in higher
prevalence in patients with periodontal disease. In addition, further analysis into its genome
indicate genes that encode for potentially virulent genes, including pro-inflammatory genes such
as cytotoxic necrotizing factor 1 and hemolysin toxic protein as well as type III secretion protein,
which could allow it to evade the immune response. qPCR analysis has demonstrated that
Saccharibacteria c​ ould inhibit an immune response against its host through the inhibition of the
TNF-ɑ cytokine production by macrophages. These findings suggest but do not confirm a link
between ​Saccharibacteria​ and pathogenicity.

The bacteria-to-bacteria interactions between ​Saccharibacteria​ and its host are quite rare, given
the diversity of the bacterial kingdom. Recent studies, observing the parasite and host
relationship under the microscope, indicate that the parasitic effect on the host is amplified in
oxygen- and nutrient-deprived conditions, evident by the upregulation of stress proteins,
hindered growth and changed the morphology of the host, ​Actinomyces odontolyticus​.

The study of ​Saccharibacteria​, its host and their interactions is vital in understanding their roles
in periodontal disease, the mechanisms of parasitic bacteria interactions and for the methods of
isolating and cultivating other uncultivable bacteria.
 Lin 6

Objectives
In order to further understand bacterial behavior and its interactions with its environment, it is
important to be able to quantify and to measure bacterial growth. Due to its minute size and
inability to be cultured independently, ​Saccharibacteria​ evades traditional quantification
techniques, such as optical density and colony quantification through plating. This study seeks to
resolve these issues through a unique molecular approach using quantitative polymerase chain
reaction (qPCR). In addition, the developed assay will also be used to measure the bacterial
growth curve of ​Saccharibacteria​ in relation to its host.

Criteria
In order to develop a rigorous quantification assay, certain parameters must be fulfilled. The
following parameters were compiled using the guidelines in the Food and Drug Administration
(FDA):
1. Specificity - the primers must demonstrate the ability to bind specifically to its region of
DNA
2. Linearity - the regression line of the assay should be analyzed through the r2 coefficient
3. Precision - the assay must demonstrate the ability to reproduce data in replicates and
under multiple concentrations (minimum: 3)
4. Range - the assay must display the ability to detect the substance over a multitude of
concentrations; the assay must also demonstrate a lower and upper limit to detection
5. Robustness - the data in the assay must display the ability to be reproduced under certain
impurities
 Lin 7

Assay Development and Methods

Primer Design
The genomes of species in the ​Saccharibacteria​ phylum were compiled from the Human Oral
Microbiome Database. Conserved regions were identified, and primers were designed after
fulfilling two criteria: conserved regions must be no more than 20 bp in length and the conserved
regions must be no more than 200 bp apart. Conserved regions and genomes of other phyla were
compared to ensure specificity.

Primer Optimization
A gradient PCR was performed using the DNA, extracted from ​Saccharibacteria​ and ​P.
propionicum​ coculture and the designed primers varying with a range of annealing temperatures
from 49.8°C to 65.1°C, determined by the machine. Gel electrophoresis was performed
afterward of each temperature sample.

Plasmid
Plasmid pJL02 was designed using a TOPO vector backbone and the conserved region. pJL02
was transformed into ​E. coli​ on LB agar plates with 20 µL/mL kanamycin and incubated for
24hrs at 35°C. Colonies in the LB agar were screened for successful transformation and were
grown in TSBY broth with 20 µL/mL kanamycin. pJL02 was isolated and sequenced to verify
that pJL02 contains the desired region of DNA with BLAST.

qPCR
After plasmid isolation, calculate the concentration of pJL02 using the Qubit Fluorometer.

Using the DNA concentration, the genomic equivalent of the undiluted pJL02 stock was
calculated. Afterward, calculated genomic equivalent to dilute was 10​8​ copies in 50 µL. A serial
dilution of the plasmid was performed from 10​7​ to 10​1​ copies of the plasmid. qPCR was run
using replicates of each dilution as template DNA as well as a negative control to detect any
contamination.

Application
Three cultures of AC001 were incubated at 37°C for 0, 8, 16, 24, 32 and 40 hrs. After each time
frame, the optical density was calculated to measure host growth and extract DNA from 1 mL of
each culture afterward. The concentration of DNA was evaluated for each of the three cultures; it
was diluted to 1ng/µL in 50 µL and diluted again using serial dilutions to the fifth dilution factor.
qPCR was performed using the template DNA and dilutions of each culture in triplicates, and
negative control was included to detect contamination. The process was repeated for each time
course.
 Lin 8

Results

Primer Design and Optimization

Identification of Specific, Conserved Regions

Figure 1: Sequence alignment using Geneius software of 14 ​Saccharibacteria​ species (left, forward primer and right, reverse primer)
A sequence alignment of 14 species in the ​Saccharibacteria​ identified two conserved regions
that fulfilled the criteria for primers. The forward primer is 19 bp long and the reverse primer is
18 bp long; these parameters satisfy the 20-25 bp upper limit for primers. In addition, the
distance between these conserved regions was 110 bp, which satisfies the 100-200 bp upper
limit.
Primer Sequence

AJ12 (Forward Primer) 5’ - GTGACTGGGCGTAAAGAGTT - 3’ (685-704)

AJ13 (Reverse Primer) 5’ - GTAGGAGTGAAATCCGTAG - 3’ (814-832)

Verification of Primer Specificity to Saccharibacteria Phylum

Figure 2: Sequence alignment of conserved ​Saccharibacteria​ regions and genomes of other oral bacteria in other phyla.
To confirm primers are specific exclusively to the ​Saccharibacteria​ phylum, the primer
sequences were compared with the species of other oral bacteria of different phyla. The sequence
alignment revealed multiple mismatches between the primers and other bacteria, which
confirmed their specificity.
 Lin 9

Figure 3: BLAST results of pJL02 DNA and ​Saccharibacteria​ HMT-488 DNA, the specific species that we selected for cloning.
An analysis of the plasmid, pJL02, which contained the PCR product annealed by the primers,
was performed to further verify primer specificity. The DNA sample made a 100% match to its
Saccharibacteria ​species.

Primer Optimization

Figure 4: Gel electrophoresis of the gradient PCR using the newly-designed primers and ​Saccharibacteria​ DNA.
The gel electrophoresis of the gradient PCR indicates that the most luminous bar at the correct bp
of 100-200 bp, which indicates the temperature where the DNA was most amplified, was 54.3
degrees Celsius. The optimal annealing temperature is important to ensure that the temperature is
not too low that the primers lose specificity and not too high that the primers will not bind.
 Lin 10

qPCR
Standard Curve and Efficiency

After performing the qPCR, the threshold

cycles (Ct) and compared to its dilution factor.
Using the qPCR data, a regression curve was
compiled. The r2 value indicates the accuracy
of the dilutions; an acceptable r2 value ranges
from 0.90 and 1. Since the regression curve for
the equation was 0.997, our dilutions and data
can be interpreted as accurate. Next, it is also
important to measure primer efficiency to
ensure that primers are amplifying the correct
region at the correct rate (double the amount of
DNA for each subsequent cycle) and that
dilutions and pipetting are correct. The
efficiency of the primers is based on the
proximity slope in the regression line to -3.32.
Efficiency is measured as a percentage from
−1
the formula: E= 10( slope ) − 1 . After plugging in
the slope of the regression equation (-3.5473)
into the formula, the efficiency of our reaction
is 91.38%. An acceptable range of efficiency is
90%-110%, so our calibration lines fall
comfortably within those parameters.
 Lin 11

Primer Robustness Verification

Since the plasmid was made using DNA from a coculture of both ​P. propionicum a​ nd
Saccharibacteria​, it is important to verify once more that the primers bind specifically to only
Saccharibacteria​ DNA per criteria 5 from the FDA assay verification criteria. To do this,
multiple dilutions of ​P. propionicum​ DNA were added, in triplicate, to 107 genomic copies. The
Ct values of each dilution were compiled and no significant changes ( C t(107 ) ± 1 ) in Ct value
were observed relative to a normal sample of 107 genomic copies without host DNA. This
indicates that the primers bind specifically to ​Saccharibacteria​ and an increase in host DNA will
not affect the results.
Figure 6: Analysis of the Ct curves to ensure primer specificity.
Blue indicates the average with the standard deviation of 1 Ct
cycle. Green depicts the 10ng of host DNA added to 107 plasmid
copies. Red depicts 1ng of host DNA added to 107 plasmid
copies. Orange depicts 0.1ng of host DNA added to 107 plasmid
copies. No significant changes observed when adding ​P.
propionicum D​ NA.

Application
The growth of infected and uninfected ​P. propionicum​ was obtained using a spectrophotometer
and measured in OD600 . Saccharibacteria​ was compiled using the developed and verified qPCR
assay. The development of the first known growth curve for ​Saccharibacteria​ reveals much
information about the species. First, it indicates that the bacteria grow at a similar rate as its host
through the stages of lag, log, stationary and death. This further suggests that ​Saccharibacteria
requires a ​living​ host to survive. Another significant detail to note is the sudden drop of
Saccharibacteria ​between 24 and 32 hours. This indicates the possibility of early onset of
stationary phase in order to conserve resources as its host also goes into the stationary phase.

Figure 7: Growth curve of uninfected ​P. propionicum,​ infected ​P.

propionicum​ and ​Saccharibacteria growth.​
 Lin 12

Discussion
A novel method for quantifying ​Saccharibacteria​ without the need for a pure culture was
developed. The process of developing a qPCR assay to quantify ​Saccharibacteria​ involved three
steps: 1) primer design and optimization, 2) plasmid design, transformation and isolation and 3)
the actual qPCR and building of the standard curve. In the first step, a gradient PCR and agarose
gel electrophoresis indicated that the optimal annealing temperature is 54.3°C. The plasmid was
designed and transformed into ​E. coli;​ it was later screened for transformation success, isolated
from the ​E. coli​ and sequenced to verify its sequence with ​Saccharibacteria​ HMT-488 using
BLAST. Finally, qPCR was conducted with a specific number of plasmid copies diluted from
10​8​ copies. After efficiency was verified (91.38%), a standard curve was developed to accurately
quantify ​Saccharibacteria​. After verifying that the primers bind to the correct region using
sequencing and BLAST, that the regression line had an appropriate r2 value, that the data can be
reproduced in triplicates, that the assay had a wide range of sensitivity and that additional host
DNA does not affect the end result, the assay meets all five parameters of assay development.

The growth of ​Saccharibacteria H ​ MT-488 was measured using the developed qPCR assay in
addition, the growth of ​P. Propionicum ​in an uninfected culture and in a culture infected with
Saccharibacteria​ HMT-488. As expected, the growth between parasite and host are similar for
the first two stages: 1) lag phase begins at 0 hours and ends at 8 hours and 2) growth phase
begins at 8 hours and ends at 24 hours. Stationary phase seems to begin earlier in the 24- to
32-hour time frame in ​Saccharibacteria​ HMT-488 than in ​P. Propionicum.​ This indicates that
Saccharibacteria​ may sense the preparation for the onset of stationary phase in its host –
potentially through molecules released by the host – and enter stationary phase first to conserve
resources.

Further investigation, including the inclusion of a time point between 24 and 32 hours, is needed
to provide greater insight into this hypothesis. Future works including improve the resolution of
growth curves by shortening intervals, understanding ​Saccharibacteria​ behavior by quantifying
its growth under various conditions or developing a risk assessment tool for periodontitis based
on differing levels of ​Saccharibacteria​.
 Lin 13

References
He, X., Mclean, J. S., Edlund, A., Yooseph, S., Hall, A. P., Liu, S., . . . Shi, W. (2014).
Cultivation of a human-associated TM7 phylotype reveals a reduced genome and
epibiotic parasitic lifestyle. Proceedings of the National Academy of Sciences, 112(1),
244-249. doi:10.1073/pnas.1419038112

Bor, B., Poweleit, N., Bois, J. S., Cen, L., Bedree, J. K., Zhou, Z. H., . . . Shi, W. (2015).
Phenotypic and Physiological Characterization of the Epibiotic Interaction Between
TM7x and Its Basibiont Actinomyces. Microbial Ecology, 71(1), 243-255.
doi:10.1007/s00248-015-0711-7

Trung, T. T., Hetzer, A., Göhler, A., Topfstedt, E., Wuthiekanun, V., Limmathurotsakul, D., . . .
Steinmetz, I. (2011). Highly Sensitive Direct Detection and Quantification of
Burkholderia pseudomallei Bacteria in Environmental Soil Samples by Using Real-Time
PCR. Applied and Environmental Microbiology, 77(18), 6486-6494.
doi:10.1128/aem.00735-11

Yang, Y., Chen, M., Yang, B., Huang, X., Zhang, X., He, L., . . . Hua, Z. (2015). Use of 16S
rRNA Gene-Targeted Group-Specific Primers for Real-Time PCR Analysis of
Predominant Bacteria in Mouse Feces. Applied and Environmental Microbiology, 81(19),
6749-6756. doi:10.1128/aem.01906-15

Cirelli, T., Finoti, L. S., Corbi, S. C., Anovazzi, G., Nepomuceno, R., Orrico, S. R., . . .
Scarel-Caminaga, R. M. (2017). Absolute quantification of Aggregatibacter
actinomycetemcomitans in patients carrying haplotypes associated with susceptibility to
chronic periodontitis: Multifaceted evaluation with periodontitis covariants. Pathogens
and Disease, 75(7). doi:10.1093/femspd/ftx092

Center for Drug Evaluation and Research. (2018). ​Bioanalytical Method Validation: Guidance
for Industry​ (USA, Department of Health and Human Services, FDA). Silver Spring, MD

Dewhirst, F. E., Chen, T., Izard, J., Paster, B. J., Tanner, A. C., Yu, W. H., Lakshmanan, A., …
Wade, W. G. (2010). The human oral microbiome. ​Journal of Bacteriology​, ​192​(19),
5002-17. doi: 10.1128/JB.00542-10

Bor, B., Mclean, J. S., Foster, K. R., Cen, L., To, T. T., Serrato-Guillen, A., . . . He, X. (2018).
Rapid evolution of decreased host susceptibility drives a stable relationship between
ultrasmall parasite TM7x and its bacterial host. ​Proceedings of the National Academy of
Sciences,115​(48), 12277-12282. doi:10.1073/pnas.1810625115
 Lin 14

Appendices
A. Plasmid Sequencing Data

B. qPCR

Template DNA Triplicate No. 1 Triplicate No. 2 Triplicate No. 3

(genetic copies) Ct values Ct values Ct values

10​7 16.319 16.522 15.771

10​6 20.050 19.721 19.715

10​5 24.897 23.519 23.158

10​4 28.035 26.728 26.973

10​3 30.483 30.024 29.502

10​2 33.418 33.412 33.476

10​1 33.440 32.895 34.738

NTC 34.095 N/A N/A

 Lin 15

C. Growth Curve Data

Hour Uninfected ​P. propionicum Infected ​P. propionicum

0 0.049333 0.054

8 0.089 0.136

16 0.415333 0.408667

24 1.14 0.966667

32 1.596667 1.113333

40 1.69 1.14

50 1.806667 1.156667