n vitro bioassembled human extracellular matrix and its application in human embryonic stem cell cultivation 6

3.7.2 Morphology of hESCs Phase contrast pictures of hESCs cultured on Matrigel, DxSDOC and DxSDOCDOC matrices up to passage 5 and passage 20 are shown in Figure 7B.. 3.7.3 Pluripotency

In the initial 5 passages, hESCs on the three matrices grew at a similar rate, as evident from the similar gradients on the population doublings graph Beyond the 5th passage, hESCs on Matrigel began to slow down their population doubling rates, with a gentler slope hESCs on DxSDOC and DxSDOCDOC continued with the same steady gradient with no apparent loss of doubling rate, finally finishing with 19.9% and 16.8% more than the population

In agreement with the observations in section 3.5.1 and 3.6.1, DxSDOC and DxSDOCDOC matrices using the enzyme dispase again allowed for higher population doublings of hESCs compared to the control

3.7.2 Morphology of hESCs

Phase contrast pictures of hESCs cultured on Matrigel, DxSDOC and DxSDOCDOC matrices up to passage 5 and passage 20 are shown in Figure 7B Arrows indicate areas of spontaneous differentiation

At passage 5, hESCs on Matrigel, DxSDOC and DxSDOCDOC grew in tight round colonies The cells within the colonies were small, with high nucleus to cytoplasm ratios and prominent nucleoli The colonies had distinct edges for each sample There were a few areas of differentiation (arrows) for the hESCs

on DxSDOCDOC but they were a small minority of the entire population

At passage 20, hESCs on Matrigel grew in smaller colonies with diffuse borders and the periphery cells were large and flattened There were widespread areas of differentiation Any pluripotent colonies were almost non-distinguishable, only recognizable by the difference in the sizes of colony cells compared to the differentiated cells

hESCs on DxSDOC grew in large round colonies that still had a dense morphology The colonies have distinct borders and the cells within were small with prominent nucleoli and high cytoplasm to nucleus ratios There

were however, some areas of spontaneous differentiation at the periphery, albeit less than that seen in the control hESCs on Matrigel The areas of differentiation had large, flattened cells, indicated by arrows

In contrast, hESCs on DxSDOCDOC grew in large round colonies that still remained dense The colony cells were also small and tightly packed, with clear distinct colony edges There were areas of differentiation at the colony periphery but these were far less than that seen in the control cells and the hESCs cultured on DxSDOC

3.7.3 Pluripotency markers levels by flow cytometry

A panel of five pluripotency markers, Oct-4, SSEA-3, SSEA-4, TRA-1-60 and TRA-1-81, were used to characterize the hESCs after 5 passages on either Matrigel, DxSDOC or DxSDOCDOC matrices For each hESC sample, triplicates of 10,000 labeled cells were used for each marker An average of the percentage of positively-labeled cells of each triplicates was plotted, together with the standard deviation (Figure 7C)

Oct-4 levels expressed by hESCs on Matrigel were at 58.9% and the levels produced by hESCs on DxSDOC and DxSDOCDOC were decidedly lower at 11.3% and 38.1% SSEA-3 marker levels were similar in hESCs on Matrigel (61.3%) and on DxSDOC (64.4%), but were higher in hESCs on DxSDOCDOC (69.3%) SSEA-4 levels for hESCs on Matrigel and DxSDOC were both similar at 62.9% and 63% respectively, but were higher again for

Matrigel at 31.2% and comparatively higher for hESCs on DxSDOC (39.6%) and DxSDOCDOC at 45% TRA-1-81 levels were not significantly different

in hESCs on Matrigel (38.2%), DxSDOC (43.8%) and DxSDOCDOC (53.7%)

Hence, hESCs on DxSDOCDOC scored better in the markers 3,

SSEA-4 and TRA-1-60 compared to hESCs on Matrigel, but were lower for Oct-SSEA-4

The same panel of five pluripotency markers, Oct-4, SSEA-3, SSEA-4, 1-60 and TRA-1-81 was used to characterize the hESCs after 20 passages on either Matrigel, DxSDOC or DxSDOCDOC matrices Again, triplicates of 10,000 labeled cells were used for each marker An average of the percentage

TRA-of positively labeled cells TRA-of each triplicate was plotted, together with the standard deviation (Figure 7D)

It was noted that hESCs on Matrigel and DxSDOC had relatively higher markers at passage 5 but these were decreased by passage 20 (Figure 7D)

For the intracellular pluripotency marker Oct-4, hESCs on DxSDOCDOC showed a higher percentage at 36.5%, than hESCs on Matrigel (29.7%), and DxSDOC (16.8%) It is noted that a population percentage of only 36.5% seems low, but as stated in section 3.5.3, Oct-4 is an intracellular pluripotency marker, which requires extra processing of the cell suspension to permeate the cellular membrane and nuclear membrane, thus resulting in more cells lost in the extra processing steps, which can contribute to the lower percentage seen

However, it is also important to note the relatively higher percentage of hESCs cultured on DxSDOCDOC positive for Oct-4 compared to the control hESCs

on Matrigel

Due to the limitations of using an intracellular pluripotency marker (Oct-4) the other pluripotency markers included were cell surface markers, hence preventing excessive cells lost to processing steps For surface marker SSEA-

3, hESCs on DxSDOCDOC showed a much higher population percentage at 62.8% Population percentages for hESCs on Matrigel remained low at 13.6%, and remained low also for hESCs on DxSDOC at 28.6%

For the surface marker SSEA-4, there were relatively higher population percentages for hESCs on all three matrices hESCs on DxSDOCDOC showed

a population percentage at 67.1%, hESCs on Matrigel showed 54.7% and hESCs on DxSDOC showed 74.4% Again, hESCs on DxSDOCDOC surpassed that of hESCs on Matrigel in terms of SSEA-4 expression

For the surface marker TRA-1-60, the population percentages for hESCs on Matrigel and on DxSDOC were low at 23.7% and 36.4% respectively In contrast, hESCs on DxSDOCDOC matrices had higher population percentage

at 47.5%

For the surface marker TRA-1-81, again, population percentages for hESCs on Matrigel and on DxSDOCDOC were low at 36.1% and 21.2% respectively,

while hESCs on DxSDOCDOC matrices had relatively higher population percentages at 47.1%

Hence, it is clearly evident that hESCs cultured on DxSDOCDOC matrices for

20 passages showed, on average, higher population percentages positive for the five pluripotency markers, than hESCs cultured on Matrigel for 20 passages

3.7.4 Pluripotency markers levels by immunofluorescence

At passage 5, hESCs on Matrigel, DxSDOC and DxSDOCDOC were harvested for adherent immunofluorescence analysis It was found that the hESCs under the three conditions grew in tight colonies, with the majority of cells positive for Oct-4 and SSEA-4 (Figure 7E)

When the hESCs were labeled for Tra-1-60, it was observed that the hESCs also grew in tight colonies with some areas of differentiation, indicated by arrows (Figure 7F) However, the majority of cells were positive for the pluripotency marker

Some of the hESCs on Matrigel were not positive for TRA-1-81 and SSEA-3 (arrow) (Figure 7G), however, the majority of cells under all three conditions were positive for the markers

These observations correlate to observations by flow cytometry, that the hESCs on DxSDOCDOC were not any less pluripotent than the control hESCs

on Matrigel

At passage 20, hESCs on Matrigel, DxSDOC and DxSDOCDOC were harvested for adherent immunofluorescence studies Similar to the observations in section 3.7.2, hESCs on Matrigel grew in small colonies separated by large areas of widespread differentiation These colonies remained positive for markers of pluripotency, Oct-4, SSEA-4 (Figure 7H), TRA-1-60 (Figure 7I), TRA-1-81 and SSEA-3 (Figure 7J) However, in between these small colonies, there were plenty of hESCs that were not positive for the markers, indicated by arrows

hESCs on DxSDOC fared better than the control hESCs, with the majority of cells growing in colonies that were positive for the markers (Figure 7H-J), but there were still areas of differentiation where cells did not express the pluripotency markers (arrows)

On the other hand, hESCs on DxSDOCDOC continued to grow in colonies that were positive for the pluripotency markers (Figure 7H-J), and although there were some peripheral cells that were not positive for the pluripotency markers, these cells were not as plentiful as the control hESCs or those of hESCs on DxSDOC

Hence, at passage 20, the hESCs were observed to retain a better pluripotent morphology compared to the control hESCs on Matrigel, and this observation correlates to that seen in the flow cytometry data in section 3.7.3

3.7.5 In vitro induced differentiation

hESCs were cultured on Matrigel, DxSDOC and DxSDOCDOC for 20 passages These cells were then harvested by collagenase IV and treated with the protocol stated in section 3.6.6 to induce neural differentiation All the three samples of hESCs on the different matrices managed to differentiate and the cells grew long extensions resembling axons, and were positive for neuronal marker β III tubulin (Figure 7K) This observation shows that even after 20 passages, the hESCs under all three conditions still retained their differentiation capacity to form neuronal cells

3.7.6 Karyotype

At passage 18, hESCs that were cultured on Matrigel, DxSDOC and DxSDOCDOC were harvested for karyotyping (Figure 7L) Out of a metaphase of 20, hESCs on Matrigel and DxSDOC each showed normal metaphases for 19 metaphases and 1 metaphase showed non-clonal random loss Again, as mentioned in section 3.5.6, these cells can be considered as karyotypically normal due to only 1 metaphase out of 20 having non-clonal random loss, which can be attributed to technical artifacts or random mitotic error hESCs on DxSDOCDOC retained normal karyotypes for all 20 metaphases

3.7.7 DxSDOCDOC matrices were able to maintain pluripotent hESCs

using dispase for passaging

In this propagation trial, based on population doubling observations, morphology, pluripotency marker expression, in vivo differentiation studies

and karyotype results, it was concluded once again, that hESCs cultured on DxSDOCDOC using dispase for enzymatic passaging showed superior population doubling rates while maintaining pluripotent-like morphology compared to control hESCs on Matrigel Pluripotency marker expressions either exceeded or were comparable to the control hESCs by flow cytometry and by adherent immunofluorescence studies at both early and late passages; these hESCs on DxSDOCDOC had fewer areas of differentiation compared to

the control In vitro differentiation studies showed that hESCs cultured on

DxSDOCDOC matrices retained their neural differentiation capacity after 20 passages and yet maintained their normal karyotype despite a prolonged culture period of 18 passages

3.8 Summary of Matrices that were able or unable to maintain hESC pluripotency

3.8.1 AcADOC versus DxSDOC versus DxSDOCDOC

The observation that AcADOC matrices were unable to maintain hESCs in their pluripotent state and that DxSDOC and DxSDOCDOC matrices were more successful, shows the importance of utilizing DxS in fibroblast culture MMC not only enhances collagen deposition, it also enhances the enzymatic activities of crosslinking enzymes, such as lysyl oxidase [39, 41] and transglutaminase 2 [73] The resulting overall higher degree of crosslinking could lead to a greater retention of ECM components during detergent lysis, preserving more pluripotency-promoting biochemical cues for hESCs

Yet, by comparing the more favourable results from hESCs cultured on DxSDOCDOC matrices compared to DxSDOC matrices, it can be deduced that while DOC was able to fully remove the fibroblasts, DOCDOC was able

to strip more proteins in addition to fibroblasts This extra removal of relatively more proteins results in a matrix that is poorer in comparison to DxSDOC matrices and appears to be favoured by hESCs in culture Also, the harsher DOCDOC treatment could possibly remove components that might otherwise cause the spontaneous differentiation of hESCs, hence stripping the ECM of differentiating-promoting biochemical cues

As such, the combined application of MMC and DOCDOC lysis would have resulted in an optimum ECM that has pluripotency-promoting biochemical cues while removed of differentiating-promoting cues

3.8.2 DxSDOC and DxSDOCDOC versus DxSNP40/DNase

DxSNP40/DNase was unable to allow for hESC adherence, and yet DxSDOC and DxSDOCDOC matrices were able to allow for hESC adherence and proliferation, which shows the importance of DOC treatment versus NP40/DNase treatment NP40/DNase resulted in a matrix that is far richer than that achieved by DOC (refer to section 3.3.2) It is possible that too much

of certain proteins may not necessarily be good for maintaining hESC pluripotency

3.8.3 FcNP40/DNase

The modification of protein morphology from mostly granular form to a reticular form, did not improve hESC pluripotency maintenance FcDOC matrix could be not obtained, as 6 days of Fc crowding together with the enhanced proliferation of fibroblasts resulted in so many fibroblasts that they could be not removed by DOC treatment or even DOCDOC treatment alone Nuclei staining was still visible after both treatments Moreover, cross-linking

of proteins might have become too extensive during the 6 days of crowding, which possibly resulted in the encasement of nuclei within scaffolds of the proteins, rendering these nuclei resilient to detergent treatment It is likely that the cross-linked proteins were also resilient to NP40 treatment, but the detergent was able to create areas of weakness that allowed the subsequent penetration of DNase to remove the nuclei In any case, FcNP40/DNase was a matrix that was found to be devoid of whole fibroblasts, but yet still contained

a rich network of proteins in reticular form But this matrix, despite its richness, was unable to allow hESCs to adhere, much less maintain hESCs in a pluripotent form

It is possible that the two matrices contained enough growth factors and other cytokines and the further addition of growth factors found in the culture medium, mTeSR-1, caused abnormal cell signaling for the hESCs, thereby causing their non-attachment In this case, an excessive amount of a beneficial substance might have adverse effects Indeed, it has been reported that MEF-

CM contains BMP-antagonists, noggin and gremlin, and hESCs can be maintained in the undifferentiated state on Matrigel in medium supplemented

MEF feeder layers, hESCs started expressing neuroectodermal markers instead [74], suggesting that over-supplementing noggin can also cause adverse effects on hESCs Hence, it is possible that the two NP40-treated matrices contain certain critical components such that further supplementation from the culture medium was unnecessary and detrimental

3.9 Matrices maintain iPSC pluripotency suboptimally

3.9.1 iPSCs share similarities with hESCs

iPSCs are pluripotent stem cells that are artificially derived from a differentiated adult somatic cell This transformation is achieved by inducing the ectopic expression of certain genes, usually by transfection of viral vectors iPSCs were first derived by transfection using retroviral vector expressing Oct-4, Sox2, Klf4 and c-Myc [75, 76], and the technique received much scientific attention, as it promises researchers the opportunity to obtain pluripotent stem cells, without the controversial use of embryos, and might be less prone to immuno-rejection iPSCs were found to be similar to hESCs in morphology, proliferation, expression of some pluripotency markers and the formation of teratomas [75, 76] iPSCs are typically cultured on MEFs or human fibroblasts as feeder cells or on Matrigel using MEF-conditioned medium [75, 77], similar to hESCs, and so it was postulated that iPSCs could also be maintained on DxSDOC and DxSDOCDOC matrices using defined culture medium mTeSR-1

iPSCs cultured on DxSDOC and DxSDOCDOC matrices were subcultured using dispase every 5-7 days, with control iPSCs cultured on Matrigel

3.9.2 Morphology of iPSCs on matrices DxSDOC and DxSDOCDOC

In this propagation trial, iPSCs were cultured on the matrices for up to 5 passages In contrast to hESCs, iPSCs on Matrigel did not maintain a pluripotent state after these 5 passages (Figure 8) Morphological observations showed that these control iPSCs lost their high nuclear to cytoplasmic ratio, colony sizes were very much reduced and the edges were no longer smooth and distinct These iPSCs were larger and flattened compared to hESCs grown for the same number of passages on Matrigel

iPSCs maintained on DxSDOC and DxSDOCDOC retained a more pluripotent morphology (Figure 8) These iPSCs retained their high nuclear to cytoplasmic ratio and grew in colonies with smooth, distinct edges There were some areas of differentiation where cells were larger and flattened (arrows), but these were far less than in the control iPSCs Despite the better maintenance of pluripotent morphology, the iPSCs cultured on DxSDOC and DxSDOCDOC matrices showed more areas of differentiation compared to similarly cultured hESCs