The above videos display how SynBioBot operates as a whole.
  Yet, we still need improvement in our robot movements regarding the utilization of a centrifuge in the process of subculturing. The problem is that the specifications of the centrifuge do not fit the gripper specification. Since the 15mL conical tube cab is thinner than the 30mm space between the toes of the gripper(tough PLA), it is required to be grabbed by the 10mm inner part space of the gripper(aluminum). To conduct such a gesture, the gripper should enter the internal centrifuge in a deeper manner so that there was a limitation upon the gripper entrance.
  But don’t worry! We completed the ‘get_pos’ function in Python, which contains the code that can define the conical tube’s location obtained by the vision data processing after the centrifuge. This means that we will definitely solve the problem just after the new centrifuge, which fits the current situation, arrives.
  In this regard, we are proud to say that we have achieved a Sustainable technology by ourselves with creativeness, innovativeness, and clear objectives and strategies for foundational advances to reduce human errors.

Wet Lab Result

[1] Preliminary Experiment of Lentivirus Transduction in HEK293T Cells

  We received the 𝝰5 lentivirus in February 2022 from a bio company and we conducted the transduction experiment using HEK293T cells in a level 2 biosafety cabinet. The first step was testing a range of volume or MOI (multiplicity of infection). MOI is the number of transducing lentiviral particles per cell and MOIs of 1, 5, 10 were used to determine the transduction efficiency.

  The successful infection of 𝝰5 lentivirus or 𝝰5 integrin overexpressed (OE) cell is shown by the emission of EGFP fluorescent from the infected cell (Fig 6.). Based on the result above, we decided to use MOI 10 for the further experiment.

  To select the infected cells, we used antibiotic selection method by using G418 (Geneticin). G418 is an aminoglycoside antibiotic that works by blocking polypeptide synthesis by inhibiting the elongation step in eukaryotic cells. Two days after transduction, HEK cells were incubated with 5 ug/ml G418 in DMEM medium supplemented with 1% FBS. G418 selection was conducted for 8 days with DMEM medium change interval every 2-3 days.

  After antibiotic selection for 8 days, EGFP fluorescent images showed wider distribution among the HEK cells population (Fig 7.). Some uninfected cells remained found after the antibiotic selection. To solve this issue, the selection procedure were repeated until only the 𝝰5 integrin OE HEK cells were detected. Once the stable 𝝰5 integrin OE HEK cell line generation was obtained, the expression of 𝝰5 integrin was monitored by western blot.

  The western blot profiles showed a significantly higher band intensity of 𝝰5 integrin protein observed at approximately 135 kDa compared to the control (WT) (Fig 8.)^[1]. This data indicates a successful transduction of HEK293T cells with 𝝰5 integrin lentivirus.

[2] Cell Culturing in An Open Space

  Due to the size and working space of SynBioBot, it could not be operated inside a biosafety cabinet. To improvise, we set up a simple open bench as an additional working area to prepare some of the culture work that was located in the same room with SynBioBot.

  The set up of the open bench consisted of a water bath, pipette, pipette tips, cell culture dishes, conical tubes, a tube rack and tissue (Fig 9.). The area was kept as clean as possible, and efforts were made to maintain a clean environment using clean wipers, ethanol, and gloves.

  To check whether there would be no contamination or unsuccessful cell experiment, we cultured human keratinocytes (HaCaT) cells in a clean bench and open space, simultaneously. In both conditions, cells were attached after 24 hours of incubation in 5% CO₂ and 37^oC incubator. HaCaT cells cultured in the open space were concentrated at certain areas in the petri dish and had no sign of contamination, showing the possibility to be cultured in open space.

[3] Cells Seeding and Passaging

  Cell seeding is the initial protocol step and standard procedure in cell-based experiments. For consistent experiment outcomes, a proper and standardized cell seeding process is essential. The main challenge in this step is to achieve and maintain comparable cell numbers in all replicate experiments^[2]. Here, we programmed SynBioBot to perform HaCaT cell seeding steps from taking the culture dish to storing the dish containing seeded cells in the incubator.

  We compared the time consumed by SynBioBot to conduct cell seeding to manual operation performed by a trained member of Sogang_Korea (Fig 11.). SynBioBot showed a constant operating time in cell seeding with an average time of 8:56 and SD of 0.016. Human hands were three times faster than SynBioBot with an average time of 2:52 and SD of 0.092. This result was anticipated because it took time for SynBioBot to do initial recognition and it is a single arm robot. Even though manual procedure was faster, SynBioBot showed well and constant operating steps of cell seeding.

  The cultured cells were observed for two consecutive days under light microscope 24 hours after the cell seeding. Microscope images were taken from petri dishes containing cultured HaCaT in three different areas and counted using ImageJ software. On the first day, cells counted from automated cell seeding have a slightly lower average and higher SD compared to the manual (Fig 12a.). This was possibly caused by the pipetting process of SynBioBot leaving some of the liquid containing cells. On the second day, both automated and manual had the same average number of cells and showed a growing HaCaT number. Morphologically, HaCaT cells looked healthy and there was no sign of contamination (Fig 12b.).

  After successfully programming the cell seeding using SynBioBot, we also managed a trial in cell subculturing. Passaging or subculturing is the removal of the medium and transfer of cells from a previous culture into fresh growth medium, a procedure that enables the further propagation of the cell line or cell strain^[3]. Cell passaging performed by SynBioBot incorporated phosphate buffer saline (PBS) washing, suction, trypsinization, and cell medium dividing steps.

  Cell passaging consumed a longer time compared to cell seeding as more steps were required to subculture confluent state cells. SynBioBot needed forty one minutes and fifty six seconds to perform cell passaging and the manual process needed twenty five minutes and thirty one seconds (Fig 13.). Note that centrifugation step was the issue of the automated cell subculturing as the position of the conical tube inside the centrifuge could change from the initial position after the centrifugation and SynBioBot was not able to recognize it.

[4] α5 Integrin Lentivirus Transduction in Human Keratinocytes

  Integrins are widely distributed cell surface receptors that are essential for regulating the interaction between a cell and its microenvironment to control cell fate and development^[4]. 𝝰5𝛃1 integrin binds to fibronectin and has a well-defined role in cell adhesion, migration, and matrix formation. In wound healing, human keratinocyte motility on a fibronectin substratum is critical in the re-epithelization of injured skin and in the spread of cutaneous malignancy^[5]. Knowing the pivotal role of 𝝰5𝛃1 integrin as a fibronectin receptor, we initiated to program lentivirus transduction process through SynBioBot to make 𝝰5 integrin overexpressed (OE) HaCaT cells. The 𝝰5 OE HaCaT could be potentially used to study the interaction of HaCaT and its microenvironment in healing wounds or skin tumors.

  Lentivirus transduction protocol conducted by SynBioBot was started with HaCaT cell seeding in the concentration of 3x10⁵ cells/mL. After being incubated for 18-20 hours, cells were incubated with polybrene for an hour to enhance the transduction efficiency. 𝝰5 lentivirus with MOI 10 was prepared in the antibiotic-free medium and added into the petri dish containing HaCaT cells. After overnight incubation, the virus-containing medium was replaced and continued with another overnight incubation. The lentivirus transduction process was well performed by SynBioBot in a constant time with the average of 18:40 and SD of 0.014 (Fig. 14). The manual process took the average of 5:55 and SD of 0.516.

  After transduction, HaCaT cells were observed using a confocal microscope and the number of EGFP positive cells was counted using ImageJ software. The results showed that the number of infected cells in the automated process was slightly higher compared to the manual (Fig 15a). The successfully infected HaCaT cells observed under the confocal microscope with differential interference contrast (DIC) imaging as the background, showing the EGFP expression that followed the cell shape (Fig 15b,c). These results revealed the capability of SynBioBot to perform lentivirus transduction as part of synthetic biology research.