Episomal vector reprogramming efficiency in human dermal fibroblasts

Answer

No single source-reported figure answers the question: episomal (oriP/EBNA-1) plasmid efficiency in human fibroblasts is printed as percentages from 0.003% to 0.1%-0.2% and as colony counts from about 3 to about 1000 per 10^6 input cells, depending on cells, factor set, matrix and counting rule. For the standard pCXLE (OCT3/4, SOX2, KLF4, LIN28, L-MYC plus p53 shRNA) and Epi5-type sets the percentages cluster low: a bank-wide mean of 0.017% (SD 0.027, median 0.006) over 49 lines, colonies per somatic cell seeded 04df8b73 8d4a2367; a mean of 0.013% over successfully reprogrammed samples b537b99e; 0.011% in neonatal BJ and 0.003% in adult scar-tissue fibroblasts, AP-positive colonies per cell plated 0aa26a97 48226595; and 0.05% in BJ c6e81ca4. Optimised protocols print higher: ~220 iPSC colonies from 1x10^6 input fibroblasts feeder-free with small molecules fb4aca52 and about 1000 per 10^6 when L-MYC replaces c-MYC c0f671fd.

Summary

Several papers are extracted more than once (Yu 2011 four times, Drozd, Jaffer and Kuebler twice), so the number of independent results is smaller than the number of facts.

Three measurement conflicts are flagged by the platform, and all involve the Vdovin 2015 abstract-only result of about 200 colonies per 10^5 cells for episomal delivery, scored on day 21 with TRA-1-60 expression, factors Oct4, Sox2, Klf4, L-Myc and Lin28 66dc8f22. That count is set against Schlaeger 2015, 0.013% b537b99e, and Trevisan 2017, 0.05% c6e81ca4; the third flagged pair is Schlaeger 0.013% against Trevisan 0.05%. The corpus does not settle any of them: the sources differ in the printed unit, in what the denominator is, in the cells and in the culture (details in the Evidence section).

Within single papers, moving from feeders to a feeder-free defined matrix raised the printed figure in every case the corpus holds: ~3 iPSC colonies from ~1x10^6 versus ~220 iPSC colonies from 1x10^6 input fibroblasts fb4aca52 c0f671fd, 0.012% versus 0.103% 616a0b3f, and 0.02% versus 8.55% marker-positive cells by flow cytometry 9a2d5ec9.

Evidence

Percentages for plasmid sets, with the factor set recorded

• Kuebler et al. 2025 - three OriP/EBNA1 vectors (OCT3/4, SOX2, KLF4, Lin28, L-Myc, p53 shRNA; Addgene 27080, 27078, 27077), 5.0x10^5 cells nucleofected, replated on day 7 onto Matrigel in ReproTeSR; mean 0.017% (SD 0.027, median 0.006); first colonies about 16 days after nucleofection; counting day and marker not stated (colonies judged by morphology); oxygen not stated for the episomal arm. 04df8b73 8d4a2367
• Schlaeger et al. 2015 - same pCXLE factors with shP53, nucleofected fibroblasts on MEFs, split onto two 10-cm plates around day 6; 0.013% as the geometric mean of successfully reprogrammed samples, colonies per somatic input cell, Epi success rate 93%; counting day not stated; oxygen not stated for Epi. The record (raw_data.finding) quotes the paper: 'The mean efficiency of successful reprogramming experiments was 2.1% for the mRNA method (n = 3 successfully reprogrammed samples), followed by SeV (0.077%) and Epi (0.013%) reprogramming'. b537b99e
• Drozd et al. 2015 - Epi5-type set (OCT3/4, SOX2, KLF4, L-MYC, LIN28, mp53DD, EBNA-1), 8x10^4 cells per well on Geltrex; AP-positive colonies per total cells plated; 0.011% in neonatal BJ (day 21) and 0.003% in adult scar-tissue fibroblasts (day 28); the paper summarises fibroblasts as approaching 0.006%; replating and reprogramming-phase oxygen not stated. 0aa26a97 48226595
• Manzini et al. 2015 - pCXLE set plus EBNA1, 6x10^5 fibroblasts per Nucleofector transfection, MEF feeders after day-7 replating; 0.017%, 0.0166% and 0.003% for three patients, nascent colonies per cell transfected; scoring day and marker not stated; Neon electroporation gave far lower values. 0a82bbda
• Trevisan et al. 2017 - BJ neonatal fibroblasts, 5x10^5 cells nucleofected with pCXLE-hOCT3/4-shp53-F, -hSK, -hUL, day-7 reseed onto MEFs; about 0.05% as TRA-1-60-positive colonies over starting cells, shared with Sendai in the same sentence; counting day, oxygen and reseeded fraction not stated. c6e81ca4
• Goh et al. 2013 - BJ cells, Yamanaka set (Oct4, Sox2, Lin28, L-Myc, p53 shRNA); 0.103 +/- 0.008% on Matrigel (76.33 +/- 5.6 TRA-1-81-positive colonies, n=3) versus 0.012 +/- 0.008% on feeders; scored day 30; denominator is transfected cells (GFP transfection efficiency approximately 36+-16%); oxygen not stated. 616a0b3f
• Thornton et al. 2021 - normal adult dermal fibroblasts, OriP/EBNA1 set with 4.5x10^5 cells nucleofected, iMEF feeders; AP-positive primary colonies at day 21, efficiency 0.1%-0.2% (the 0.15 recorded in the measurement is only the midpoint); seeded number per well and oxygen not stated. 35b3dedf
• Hu et al. 2016 - adult dermal fibroblasts, episomal plasmids with small molecules, xeno-free on vitronectin; approximately 100 colonies from 1x10^5 cells, about 0.1%; abstract only, so factor set, day and marker are not stated. a628a559

Counts per input cells

• Vdovin et al. 2015 - adult human skin fibroblasts, oriP/EBNA-1 vectors with Oct4, Sox2, Klf4, L-Myc, Lin28; about 200 colonies per 10^5 cells on day 21, colonies expressing TRA-1-60, versus 8 for Sendai; abstract only; whether 10^5 is plated or transfected cells, feeder or matrix, replating and oxygen are not stated. 66dc8f22
• Yu et al. 2009 - newborn foreskin fibroblasts, seven-factor set (OCT4, SOX2, NANOG, LIN28, c-Myc, KLF4, SV40 large T); about 3 to 6 colonies per 10^6 input cells; counting day and marker not stated. d1e1e190
• Yu et al. 2011 - same seven-factor class, nucleofected foreskin fibroblasts; original MEF-feeder method about 3 colonies from about 1x10^6, feeder-free small-molecule method about 220 colonies from 1x10^6 (about 70-fold, AP-positive, day 22, from about 0.33x10^6 input cells per dish), and about 1000 per 10^6 with L-MYC replacing c-MYC (counted day 21, n=3); oxygen not stated. fb4aca52 a3ee4523 f6dbd070 c0f671fd
• Chen et al. 2011 - episomal combination #19 plus EBNA mRNA in E8 on vitronectin; efficiencies ranging from four to 200 iPS cell colonies per 10^6 fibroblasts initially transfected for banked lines and 60 to ~1,000 iPS cell colonies per 10^6 transfected fibroblasts for biopsy-derived fibroblasts; counted without passage; counting marker not stated; the default condition for most experiments is hypoxic (5% O2, 10% CO2). 92476c53
• Hung et al. 2016 - BJ, OCT4, SOX2, KLF4, L-MYC, LIN28 and p53 shRNA, vitronectin, day 28; 37 +/- 6 TRA-1-60-positive colonies per 50,000 reprogrammed cells (n=3); oxygen and replating not stated. ca8bf611
• Jaffer et al. 2018 - AP-positive colony counts at day 25-30, no denominator printed: BJ 200 +/- 52.9, adult line PD1 65 +/- 1.5 (70 +/- 2.0 in a second panel), RPD1 24 +/- 4.2; GFP transfection efficiency 32.8 +/- 1.68%. 12cd4282 c3109453
• Howden et al. 2015 - patient fibroblasts, seven-factor set plus miR-302/367 vector; more than 1,000 colonies per electroporation experiment with no per-cell denominator, and more than 100-fold more colonies with the miR vector. 16a47724

Not an efficiency per cell, or not a number

• Valamehr et al. 2014 - FTC007 fibroblasts (dermal origin not stated), seven-factor set; 8.55% SSEA4/TRA181/CD30-positive cells by flow cytometry in feeder-free FRM/FMM versus 0.02% in conventional medium on feeders; a marker-positive fraction, not colonies per cell plated. 9a2d5ec9
• Ali et al. 2023 - dermal fibroblasts, three pCXLE plasmids; reports success per reprogramming attempt (100%, 50%, 33%) with first colonies on days 28-33, not colonies per cell; no structured measurement. 7c9c8906
• Bang et al. 2018 - BJ cells, day-14 AP colonies; optimum 3 ug of each of three vectors and 1.0x10^5 cells per well; no numeric value is printed in the text (bar charts only). 56835311

The three platform-flagged conflicts

| Left | Right | Why this is not a same-condition conflict | |---|---|---| | about 200 colonies per 10^5 cells 66dc8f22 | 0.013% b537b99e | Vdovin is an abstract-only count with an unstated denominator type, matrix and oxygen; Schlaeger is a geometric mean over successful samples only, on MEFs after a day-6 split. The units differ and no conversion is made here. | | about 200 colonies per 10^5 cells 66dc8f22 | 0.05% c6e81ca4 | Adult skin fibroblasts, TRA-1-60 expression, day 21 versus neonatal BJ, TRA-1-60 over starting cells, no counting day; the factor sets are different too (no p53 suppression named in the Vdovin abstract). | | 0.013% b537b99e | 0.05% c6e81ca4 | Both use pCXLE nucleofection onto MEFs, but one is a success-conditioned mean over fibroblast samples with a day-6 split to two 10-cm plates and the other a single BJ line with a day-7 reseed; neither states a counting day. This is the closest pair, and the corpus does not settle whether the difference is biological. |

Methodology

Every verified fact supplied for this task is used. Percentages, counts per defined number of cells, a flow-cytometry fraction and per-attempt success rates are kept as separate quantities, and none was converted into another unit. Repeated extractions of one paper are grouped, not counted as replications. Factor sets are recorded because Epi5 (mp53DD plus EBNA1), pCXLE (p53 shRNA, L-MYC, LIN28) and the seven-factor Yu sets are not interchangeable with each other or with OSKM Sendai kits. Where a record says dermal origin is not stated (Hu excepted, Valamehr, Howden, Jaffer for BJ), that is carried forward.

Limitations

• Few facts share cell type, factor set, matrix, counting day and denominator, so the corpus gives no matched cross-paper estimate; neonatal BJ lines dominate and adult dermal biopsy data are sparse 0aa26a97 35b3dedf a628a559.
• Oxygen is stated for the episomal reprogramming phase in almost none of the facts; only Chen gives a default hypoxic condition 92476c53.
• Counting day and colony marker are often unstated or differ (AP, TRA-1-60, TRA-1-81, morphology) 04df8b73 0a82bbda 0aa26a97 616a0b3f.
• Two facts rest on abstracts alone 66dc8f22 a628a559; several print counts with no denominator 12cd4282 c3109453 16a47724; Ali and Bang give no usable efficiency number 7c9c8906 56835311.
• Yu 2009 records a lower bound of the printed range as its value d1e1e190, and Thornton records a midpoint of a printed range 35b3dedf; both ranges are quoted as printed above.