Antibodies

Origin of antisperm antibodies

During the 20th century, several cases of infertile men and women were observed. At this time, the reasons of this sterility were elusive. Researchers discovered that this was the result of the production of antisperm antibodies (ASA). These antibodies can be produced in the body of women and men against spermatozoa, and they cause infertility. Presence of ASA in male and/or female partner have been considered as infertility cause in around 2-30% of infertile couples [1]. For example, ASA production in women can be caused after spermatozoal deposition into genital tract with a compromised epithelial barrier, the peritoneal cavity or the gastrointestinal tract.

Researchers tried to use these antibodies and related antigens to create a contraceptive vaccine (CV) [2]. They made mice produce ASA (production by B cells) by vaccinating them with the related antigen. The results show that vaccination with a sperm antigen or its cDNA causes a long-term, reversible contraception in female mice [3].

They showed that ASA can be very effective for a long-term and reversible contraception. This is exactly what we want to create by engineering L. jensenii to make them produce ASA against a spermatozoa antigen.

Why do we chose to synthesize ASA ?

What is an antibody ?

The classical representation of an antibody is as a Y-shaped molecule composed of four polypeptide subunits with two identical heavy and light chains, linked by disulfure bonds. Each chain is composed of a constant (constant heavy CH, constant light CL) and a variable part (variable heavy VH, variable light VL). It is secreted by B lymphocytes that specifically recognize and neutralize antigens.

scFv

Single-Chain Variable Fragment (scFv) are recombinant molecules where the variable region of the heavy (VH) and light (VL) immunoglobulin chains are linked into a single peptide. VH and VL are linked with a flexible linker sequence, here Gly4Ser. Glycine amino-acids ensure a good flexibility of the linker. Effect of scFv is similar to antibodies.

Antisperm antibodies

ASA are immunoglobulins (IgG, IgA, IgM) directed against sperm antigens. The production can be caused whenever sperm encounters the immune system. They can have different effects.

Effects of ASA on sperm

ASA can affect fertility through several mechanisms : inhibition of sperm motility, capacitation, acrosome reaction, sperm zona interaction, sperm-oolemma binding and penetration and preimplantation embryonic development.
The main activity of spermicides is the inhibition of sperm motility. We chose several ASA that have this effect. If spermatozoa have their motility inhibited, they cannot fertilize the egg anymore [2].

Previous work on Lactobacillus jensenii :

An article by OSEL, a biotechnology company, discusses the production of scFv against HIV-1 virus in Lactobacillus jensenii. We studied their work in order to better understand their experiments and their constructions. OSEL defined different promoters like RpsU and PtsU that are specific to L. jensenii [5].
They determined a signal peptide that is specific to L. jensenii and added a linker to secrete antibody scFv that targets HIV-1 [4].

YLP20 and YLP12

First, we read publications to find several ASA. We found 4 interesting ASA with their related antigens.

We decided to choose YLP20 for a candidate as it is very well documented. Moreover, YLP12 - the antigen targeted by YLP20 - is small and we can easily have this peptide synthesized to test the activity of the scFv that we want to produce.

Table : ASA and related antigens.
Antibody	Related antigen
YLP20	YLP12
AFA-1	FA-1
FAB-7	FA-1
AS16	Human Sperm Extract (HSE)

YLP20

The YLP20 scFv that we want to produce is composed of a heavy chain (VH), a light chain (VL), a linker (Gly4Ser) and an E-Tag.

Role of each part :

E-Tag : purification of our scFv using anti-E-Tag antibody
linker Gly4Ser : link VH and VL
VH and VH : compose the antigen-binding domain

It is a 224 amino-acids sequence (with 13 more amino-acids for the E-tag sequence, for a total of 237) :

QVQLVESGGDLMQPGGSLRVSCAASGFTVSSSAMSWVRQAPGRGLEWVSVVYVDGTTYYADSVKGRFTISRDNSKNTLYLQMDSLTAEDTAVYYCARSNWHYVTAMYNGGGGSGGGGSGGGGSQIVLTQSPGTLSLSPGERATLSCRASQSVTMNYLAWYQQKRGQPPRLLIYAATTRATGIPDRFSGSGSGTDFTLTIRRLEPEDFAVYYCQQYGSSPPGVTFGAPVPYPDPLEPR

The bold parts are the different CDR (Coding DNA Region). The underlined part is the E-tag sequence. In blue is the light chain. In green is the heavy chain. The linker is highlighted in yellow.

YLP12

YLP12 is dodecamer, it is composed of 12 amino-acids. It is a sperm specific protein. This peptide sequence is involved in recognition and binding to the human oocyte zona pellucida (ZP), ZP protein, ZP3. YLP12 is present on acrosome, midpiece and tail regions of sperm cells.

It is a 12 amino-acids sequence : YLPVGGLRRIGG.

YLP12 was synthesized by EZBiolab with biotinylated C-terminus part.

We decided to add a biotinylated C-terminus part to fix YLP12 to use it as a capture antigen in an ELISA test. The biotinylated part is supposed to interact with the streptavidin in ELISA plate which will prevent the peptide to lie down. It will facilitate the interaction between the scFv and the peptide.

Objectives

Production

We first want to engineer L. jensenii and B. subtilis to make them produce YLP20 scFv. For that, We used a strong promoter, RpsU, that is specific to L. jensenii [5].

The scFv is synthesized from the N-terminus domain to C-terminus domain. The E-Tag is added at the end of the synthesis, on the C-term.

We used the same construction in L. jensenii and B. subtilis to test the production and secretion of our product. We do not know if the promoter works in B. subtilis as it is specific to L. jensenii, but both are gram negative bacteria. Once the scFv is fully synthesized, it is supposed to looks like Figure 1 (see below).

Figure 1 : Synthesized scFv with E-tag added at the C-terminus domain.

Secretion

The next step is to make L. jensenii and B. subtilis secrete YLP20 scFv in the supernatant using a signal peptide (SP).

Signal Peptide	Origin	Amino acid sequences
CbsA	L. jensenii	MKKNLRIVSAAAAALLAVAPVAASAVSTVSA
Epr	B. subtilis	MKNMSCKLVVSVTLFFSFLTIGPLAHA
YncM	B. subtilis	MAKPLSKGGILVKKVLIAGAVGT AVLFGTLSSGIPGLPAADA
YjfA	B. subtilis	MKRLFMKASLVLFAVVFVFAVKGAPAKA

We chose 4 differents SP to perform the secretion of scFv :

CbsA

The first that we chose is CbsA. This SP is used for secretion of antibodies in Lactobacillus jensenii [4][5].
The SP is synthesized on the N-terminus domain. Downstream the CbsA sequence are added 4 amino acids (APVT) at the N-terminus domain of the scFv. These 4 amino acids are similar to a native signal peptidase cleavage site of this protein. The addition of the APVT sequence shown to improve the secretion of a full-length protein in L. jensenii.

Epr, YncM, YjfA

These SPs are coming from a screening of SPs of Bacillus subtilis. As they can be used to secrete proteins in B. subtilis which is a gram positive bacteria, we decided to try to use them to secrete our stuff in L. jensenii that is also a gram positive bacteria. We used the same template for the construction. These SPs are on the N-terminus domain, and followed by the APVT sequence on their N-terminus domain.

Designs

The following figure (Figure 2) is the caption for the next 3 figures of our designs.

Figure 2 : Meaning of the icons for the 3 following figures.

We decided to design several constructions to produce YLP20, our favorite candidate with four differents signal peptides : CbsA, Epr, YncM and Yjfa (Figure 3). We chose RpsU promoter [5] that is a strong promoter specific to L. jensenii.
We tested different signal peptides that we introduced you before without forgetting the amino sequence “APVT” to avoid truncating the protein. Finally, we added the sequence that encodes YLP20 scFv and E-tag.

Figure 3 : Constructs that produce and secrete YLP20 with 4 different signal peptides.

Figure 4 : Constructs that produce and secrete LaM-4 (a control) with 4 different signal peptides.

To check the scFv production and conformation we designed constructions with LaM-4 instead of YLP20 sequence (Figure 4).
Indeed LaM-4 is a control [6]. It is an antibody against RFP & mcherry. This control allow us to check the efficiency of the promoter and the signal peptide.
The construction is very similar that the one with YLP20. The difference is just that we replaced the YLP20 sequence with the LaM4 sequence.

Moreover, we would like to check signal peptides that we chose. Indeed, it is really important to verify the efficiency of the SP and determine new SP that are working in L. jensenii. We decided to fusion the SP to RFP to display the efficiency of the SP (Figure 5). For that, we used a plate reader and a flow cytometer.

Figure 5 : Constructs that produce and secrete RFP with 4 different signal peptides.

Experiments

To verify if our constructs are working correctly (i.e production and secretion in the supernatant) we perform several experiments (Figure 6). We do ELISA tests using YLP12 (biotinylated as follow : YLPVGGLRRIGG-Lys(Biotin)) that is the target of the YLP20 scFv. We use streptavidin plates to fix YLP12, so the repulsion between biotin and streptavidin forces the peptides not to lie on the bottom of the plate and facilitate the binding antigen-antibody.

We perform different ELISA tests : with the supernatant of bacteria to see if YLP20 is produced and secreted as we want, and with cell lysate to see if YLP20 is just produced and not secreted. To confirm that YLP20 is well produced by the bacteria, we do a Western Blot.

Figure 6 : Proof of concept for the production and the secretion of the scFv in *Lactobacillus jensenii*. All the constructions are cloned by Gibson Assembly in *E. coli*. After checking of the constructions, *L. jensenii* is transformed by electroporation. To verify if both production and secretion of scFv are effective, a Western Blot is done. After that, an ELISA test is run to see if the produced scFv ca link to the specific peptide YLP12.

Analysis of Western Blot results

To know what is good or wrong, depending of the Western Blot results, we did a figure (Figure 7). It is supposed to help us to analyse the results that we obtained. If everything in the YLP20 sequence is working, we will have the same results in both supernatant and pellet (cell lysate) WB. If only the signal peptide is not working, we will have nothing on the WB from supernatant, and correct results with the pellet. Now, if the complexe promoter-RBS is not working (then either the promoter, the RBS, or both) and the SP is not effective, we will have nothing on the WB from the supernatant and from the pellet.

Figure 7 : Analysis of the WB results depending of what works and what does not work in the sequence.

References
[1]	Rajesh K. Naz, Subhash C.Chauhan. 2001. Presence of antibodies to sperm YLP12 synthetic peptide in sera and seminal plasma of immunoinfertile men. Molecular Human Reproduction Vol.7 no.1 pp. 21–26.
[2]	Rajesh K. Naz. 2014. Vaccine for human contraception targeting sperm Izumo protein and YLP12 dodecamer peptide. Protein Science 2014 Vol.23:857—868.
[3]	A.S. Samuel and R.K. Naz. 2008. Isolation of human single chain variable fragment antibodies against specific sperm antigens for immunocontraceptive development. Human Reproduction Vol.23, No.6 pp. 1324–1337.
[4]	Angela Marcobal, Xiaowen Liu, Wenlei Zhang, Antony S. Dimitrov, Letong Jia, Peter P. Lee, Timothy R. Fouts, Thomas P. Parks, and Laurel A. Lagenaur. 2016. Expression of Human Immunodeficiency Virus Type 1 Neutralizing Antibody Fragments Using Human Vaginal Lactobacillus. Aids Resaerch And Human Retroviruses Volume 32, Number 10/11.
[5]	Xiaowen Liu, Laurel A. Lagenaur, David A. Simpson, Kirsten P. Essenmacher, Courtney L. Frazier-Parker, Yang Liu, Daniel Tsai, Srinivas S. Rao, Dean H. Hamer, Thomas P. Parks, Peter P. Lee and Qiang Xu. 2006. Engineered Vaginal Lactobacillus Strain for Mucosal Delivery of the Human Immunodeficiency Virus Inhibitor Cyanovirin-N. Antimicrob Agents Chemother Vol. 50, No. 10, p. 3250–3259
[6]	Fridy, P. C., Li, Y., Keegan, S., Thompson, M. K., Nudelman, I., Scheid, J. F., ... & Rout, M. P. 2014. A robust pipeline for rapid production of versatile nanobody repertoires. Nature methods, 11(12), 1253.

Team:Montpellier/Antisperm/antibodies