Truncated vitronectin with E-cadherin enables the xeno-free derivation of human embryonic stem cells

Human embryonic stem cells (hESCs) have unique abilities that enable their use in cell therapy, disease modeling, and drug development. Their derivation is usually performed using a feeder layer, which is undefined and can potentially cause a contamination by xeno components, therefore there is a tendency to replace feeders with xeno-free defined substrates in recent years. Three hESC lines were successfully derived on the vitronectin with a truncated N-terminus (VTN-N) in combination with E-cadherin in xeno-free conditions for the first time, and their undifferentiated state, hESC morphology, and standard karyotypes together with their potential to differentiate into three germ layers were confirmed. These results support the conclusion that the VTN-N/E-cadherin is a suitable substrate for the xeno-free derivation of hESCs and can be used for the derivation of hESCs according to good manufacturing practices.


Truncated vitronectin with E-cadherin enables the xeno-free derivation of human embryonic stem cells
Tereza Souralova 1,2 , Daniela Hulinova 1,2,3 , Michal Jeseta 4 , Pavel Ventruba 4 , Ales Hampl 1,5 & Irena Koutna 1,2* Human embryonic stem cells (hESCs) have unique abilities that enable their use in cell therapy, disease modeling, and drug development.Their derivation is usually performed using a feeder layer, which is undefined and can potentially cause a contamination by xeno components, therefore there is a tendency to replace feeders with xeno-free defined substrates in recent years.Three hESC lines were successfully derived on the vitronectin with a truncated N-terminus (VTN-N) in combination with E-cadherin in xeno-free conditions for the first time, and their undifferentiated state, hESC morphology, and standard karyotypes together with their potential to differentiate into three germ layers were confirmed.These results support the conclusion that the VTN-N/E-cadherin is a suitable substrate for the xeno-free derivation of hESCs and can be used for the derivation of hESCs according to good manufacturing practices.
Human embryonic stem cells (hESCs) proliferate almost indefinitely without the loss of pluripotency, the potential to differentiate into any cell type in a human body 1 .Self-renewal and pluripotency allow to produce high number of hESCs and subsequently hESC-derived desired cell types that can be used in regenerative therapies, drug development, and disease modeling [2][3][4] .
Regardless of the intended use of hESCs or hESC-derived cells, consistent and reproducible results are desired.One factor that contributes to inconsistent results is the variability of the derivation and subsequent culture conditions.The allogeneic or xenogeneic origin of some feeders and substrates, e.g., Matrigel, mouse embryonic fibroblasts, or human foreskin fibroblasts, can be potentially the source of variability and moreover viral or bacterial contamination [5][6][7] .
The high variability of feeders and some substrates can be lowered using chemically defined recombinant xeno-free substrates.Despite the fact that defined recombinant xeno-free substrates as vitronectin, laminin 521, collagen IV, or fibronectin were used for the hESC culture [8][9][10][11][12] , only laminin 521 5 and recombinant laminin-511 E8 protein fragments 13 were used for the hESC derivation.
Vitronectin is a glycoprotein found in the extracellular matrix and blood that promotes cell adhesion and spreading 15,16 .Its truncated form VTN-N is a defined, recombinant human protein that is used for human pluripotent stem cell culture and differentiation 12,[16][17][18] .Despite its pluripotency-maintaining properties, any form of vitronectin has never been used for the derivation of hESCs according to current literature.
In this article, we describe for the first time the derivation of three hESC lines on VTN-N with attachment support provided by E-cadherin and compare their properties with the hESC lines derived on laminin 521 according to GMP that were established by our group previously 19 .

Methods
Donor testing.Embryo donors were tested for the presence of HIV1/2, hepatitis B, hepatitis C, and syphilis with negative results as previously described 19 .Testing was performed according to procedures of the CAR University Hospital Brno embryological laboratory (Brno, Czech Republic).
Embryo preparation and transport.Embryos were thawed using Warm Cleave or Warm Blast media (Vitrolife, Västra Frölunda, Sweden) at the CAR University Hospital Brno embryological laboratory (Brno, Czech Republic), a day or two days (according to frozen stages) before the transport as previously described 19 .For lower embryo stages, the cultivation to the blastocyst stage in Blastocyst Medium (cat.no.G20722, Cook Medical, Bloomington, USA) was performed.After reaching the blastocyst stage, the disruption of the zona pellucida was executed using a laser (OCTAX NaviLase).Prepared embryos in the hatched blastocyst stage were transferred to the Sydney IVF gamete buffer medium (cat.no.G48258, Cook Medical, Bloomington, USA) and transported using a temperature-controlled transport incubator at 37 °C (portable incubator, Minitube).
Derivation.The embryo derivation procedure was initiated immediately upon receipt of the embryos, following established protocols 19 .Each individual embryo was carefully introduced into a well of a 4-well dish (Thermofisher Scientific, San Jose, CA, USA) filled with pre-warmed Sydney IVF gamete buffer medium, situated on a stereomicroscope plate maintained at 37 °C.Subsequently, the embryo was delicately transferred into a small droplet of Sydney IVF gamete buffer medium (cat.no.G48258, Cook Medical, Bloomington, USA), overlaid with Sydney IVF culture oil (cat.no.G44990, Cook Medical, Bloomington, USA), using a denuding micropipette (cat.005-300-A, Microtech IVF, Czech Republic).Throughout the process, each embryo was handled separately, employing biopsy micropipettes (cat.004-35-30A, Microtech IVF, Czech Republic) and holding micropipettes (cat.001-120-30H, Microtech IVF, Czech Republic).The inner cell mass (ICM) was aspirated using a biopsy micropipette and carefully positioned in parallel with a holding micropipette, slightly overlapping to enable efficient mechanical biopsy using a rapid swinging motion (previously published video: https:// www.mdpi.com/ 1422-0067/ 23/ 20/ 12500# B41-ijms-23-12500).Subsequently, the denuding micropipette (cat.005-150-C, Microtech IVF, Czech Republic) was employed to manipulate the ICM, which was then transferred into an individual well of a 4-well dish pre-coated with 10.0 µg/mL (1.6 µg/cm 2 ) of recombinant VTN-N (cat.no.A14700, ThermoFisher Scientific), and 2.2 µg/mL (0.34 µg/cm 2 ) of E-cadherin (R&D Systems) diluted in phosphate-buffered saline (PBS).Coating of 4-well dishes with VTN-N/E-cadherin was performed for 1 h at room temperature.Throughout the entire derivation process, the microscope plates were maintained at a constant temperature of 37 °C.E-cadherin was added only for the derivation until the first passage.The derivation medium consisted of NutriStem® hPSC XF Medium (Biological Industries, Beit-Haemek, Israel), supplemented with 20 mg/mL of human serum albumin (Vitrolife), and 10 µM of the ROCK inhibitor (Y27632, GMP, Biotechne).The derivation medium was used only for the first three days of the derivation, then the NutriStem® hPSC XF Medium was changed daily.
Culture conditions.The derived hESCs were cultured under hypoxic culture conditions (5% O 2 , 5% CO 2 , 37 °C) for the first three passages, then under normoxic conditions (5% CO 2 , 37 °C) in a NutriStem® hPSC XF Medium (Biological Industries) with a daily medium change as previously described 19 .The cells were passaged mechanically by an insulin syringe (B.Braun) and cultured on 5.0 µg/mL (0.6 µg/cm 2 ) recombinant VTN-N (cat.no.A14700, ThermoFisher Scientific) for the first three passages 17 .From passage four, non-enzymatic subculturing utilizing 0.5 mM EDTA was conducted once the cell population achieved 70% confluence.To promote single cell survival and facilitate the passaging process, a ROCK inhibitor (Y27632, GMP, Bio-techne) at a concentration of 10 µM was applied, both 1 h before and immediately after the passage for the following 24 h.Following the one-hour treatment with the ROCK inhibitor, the cells were gently washed with phosphate-buffered saline (PBS, Gibco), dissociated into clumps using 0.5 mM EDTA (Invitrogen), and subsequently transferred onto culture surfaces coated with 5.0 µg/mL (0.6 µg/cm 2 ) recombinant VTN-N (cat.no.A14700, ThermoFisher Scientific).

Determination of growth curve and population doubling time.
Cells were cultured in 24-well plates at a seeding density of 2 × 10 4 cells per well and harvested between days 1 and 6 post-seeding.The viability of cells was assessed using the Countess III Automated Cell Counter (Thermo Fisher Scientific) with trypan blue exclusion.A growth curve was constructed to depict the cell population dynamics over time.The proliferation rate, expressed as Population Doubling Time (PDT), was determined at the inflection point of the growth curve.The PDT was calculated using the formula PDT = T ln(2)/ln(A/A0), where T represents the cultivation time in hours, A is the final cell number, A0 corresponds to the initial cell number and ln is natural logarithm 20 .
Immunocytochemistry.The cells were fixed using cold 4% paraformaldehyde (Sigma) for a duration of 20 min, followed by permeabilization using 0.2% Triton × (Sigma) for 30 min.Subsequently, a blocking step was performed for 1 h in a solution of 2.5% Bovine Serum Albumin (BSA, Pan Biotech) in phosphate-buffered saline (PBS, Gibco), supplemented with 0.1% Tween 20 (Sigma), as previously described 19 .For immunostaining, the fixed cells were exposed to primary antibodies overnight at 4 °C.The primary antibodies used were as follows: mouse anti-OCT3/ Karyotyping.Cells were mitotically arrested by adding 0.4 μg/mL KaryoMAX™ Colcemid™ Solution (Thermofisher) and subsequently incubated for 2 h under standard culture conditions (5% CO 2 , 37 °C) as previously described 19 .Following the detachment of cells with TrypLe express and a 25 min treatment with the hypotonic solution (DMEM/F12 with demineralized water in ratio 1:3), the cells were fixed with 4 °C methanol and acetic acid (3:1).A karyotype analysis was performed by the Cytogenetic Laboratory Brno (Brno, Czech Republic) with Giemsa-banding and microscopic examination.At least 40 metaphase spreads/samples were analyzed at a resolution of 450-500 bands/haploid set.

Results
VTN-N/E-cadherin is a suitable surface for the derivation of human embryonic stem cells.Three hESC lines, MUES 10, MUES 11, and MUES 12, were established with the use of VTN-N/Ecadherin from 20 embryos in total (see Table 1).The rate of hESC line derivation is 15.0%.Three individual derivations were performed with 3-9 embryos per derivation.Fully-hatched blastocysts with visible inner cell masses (see Fig. 1) were used for the derivation of hESC lines MUES 11 and MUES 12.The collapsed hatching blastocyst (see Fig. 1) was used for the derivation of MUES 10.The separation of inner cell mass was impossible due to its collapsed state therefore the blastocyst was split into several clumps.Inner cell masses and clumps were transferred on 10.0 µg/mL (1.6 µg/cm 2 ) VTN-N enriched with 2.2 µg/mL (0.34 µg/cm 2 ) E-cadherin.The differentiation was observed in the first passage only in MUES 11 (see Fig. 1) although the differentiation occurred in all three hESC lines in early passages until the hESC phenotype was stabilized (data not shown).

Discussion
The derivation of hESC lines on VTN-N/E-cadherin was performed with the use of hatching or fully hatched blastocysts that were thawed at center of assisted reproduction.The zona pellucida was disrupted by laser followed by transport to our laboratory.We hypothesize that the disruption of zona pellucida may cause the degradation of blastocyst during transport, therefore it would be beneficial to collect and analyze data regarding this issue.However, we achieved a 15.0% derivation success rate with the use of 20 blastocysts in total.It is difficult to   www.nature.com/scientificreports/compare the success rate as no group has derived new hESC lines on VTN-N/E-cadherin yet, but we used laminin 521/E-cadherin in our previous setting and achieved a 7.9% success rate with the use of 38 embryos in total 19 .Takada et al. derived 1 clinical-grade hESC line on recombinant laminin-511 E8 protein fragments in period one with 7.1% efficiency (14 blastocysts, 1 hESC line) which increased to 45.5% in period two (11 blastocysts, 5 hESC lines) 13 .Despite the lack of statistically sufficient amount of data, VTN-N/E-cadherin has almost two times better derivation success rate compared to derivation of new lines on laminin 521/E-cadherin by our group 19 .
Researchers in the laboratory of James Thomson found that the VTN-N supports hESC attachment and survival better than wild-type vitronectin when used in conjunction with Essential 8™ Medium 17 .Therefore we decided to use the VTN-N for the derivation instead of its wild-type form.Moreover, its supporting properties are complemented by its availability in GMP quality, so it can be potentially used in the GMP manufacture of hESCs and hESC-derived cells used in cell therapy 21,22 .Moreover, GMP VTN-N is a chemically defined xeno-free substrate that is easy to use for coating and is more cost-effective in comparison to GMP laminin 521.
The E-cadherin supports the survival, self-renewal, and pluripotent state of isolated hESCs 5,14 .We decided to double the VTN-N concentration usually used for the cell culture and add E-cadherin to support the survival of isolated inner cell mass and increase the establishment efficiency.We designed this condition based on our previous experience with the derivation of hESCs on laminin 521/E-cadherin according to GMP.
We observed characteristic hESC morphology in all three hESC lines derived on VTN-N/E-cadherin 23 .Unfortunately, it was impossible to make photos of outgrowths for MUES 10 and MUES 11 as they were on the edge of the well.We observed the occurrence of outgrowths more often on the edges of the wells in our previous derivations on laminin 521 too (data not shown).
We observed the variability among various hESC lines, derived either on laminin 521/E-cadherin or VTN-N/E-cadherin, during the embryoid bodies' development.It is difficult to make clear conclusions as three hESC lines were used, therefore there is not enough data for statistical analysis.However, it is possible these differences are caused by the biological variability of hESC lines and are independent of the derivation and culture surface 24 .Although the same number of cells was used at the beginning of the experiment, it was impossible to infer how many cells will be involved in the creation of one single EB, therefore it would be helpful to optimize the creation of EBs to make results more comparable 25 .It would be beneficial to collect more data focusing on undifferentiated state and differentiation of more hESC lines derived on VTN-N/E-cadherin and find a statistically supported conclusion if a VTN-N/E-cadherin is a superior surface for hESC derivation.

Conclusions
We used VTN-N/E-cadherin for the derivation of hESCs for the first time, derived three hESC lines, and confirmed their undifferentiated state, characteristic hESC morphology, and standard karyotypes as well as their potential to differentiate into three germ layers.All three hESC lines derived on VTN-N/E-cadherin form EBs with similar variability as hESC lines derived on laminin 521/E-cadherin.These results support the conclusion that the VTN-N in combination with E-cadherin is a suitable substrate for the derivation of hESCs.