Article Text

Download PDFPDF

Original article
The effect of gravitational force on limbal stem cell growth
  1. Ammar Miri1,2,
  2. Khurram Hashmani1,3,
  3. Muhamed Al-aqaba4,
  4. Lana A Faraj1,
  5. Usama Fares1,
  6. Ahmad Muneer Otri1,2,
  7. Dalia G Said5,
  8. Harminder Singh Dua1
  1. 1Division of Ophthalmology and Visual Sciences, University of Nottingham, Nottingham, UK
  2. 2Department of Ophthalmology, Aleppo University, Aleppo, Syria
  3. 3Department of Ophthalmology, Hashmanis Hospital, Karachi, Pakistan
  4. 4College of Medicine, University of Basrah, Basrah, Iraq
  5. 5Research institute of Ophthalmology, Cairo, Egypt
  1. Correspondence to Dr Harminder Singh Dua, Division of Ophthalmology and Visual Sciences, B Floor, Eye ENT Centre, Queens Medical Centre, University Hospital, Derby Road, Nottingham NG7 2UH, UK; harminder.dua{at}


Aim To evaluate the effect of gravity on corneal epithelial cell migration in vitro.

Methods Fourteen donor peripheral corneoscleral rims were used. Twenty explants were chosen of which 10 were placed vertically and 10 were placed horizontally during culture. Analyses were performed to investigate the effect of gravity on epithelial growth by measuring the extent of epithelial cell growth above and below the horizontal meridian and counting the total number of cells using a haemocytometer.

Results There was no statistically significant difference in cell growth between the explants that were placed horizontally and vertically. However, in the vertical explant group the cells grew preferentially towards the 6 o'clock direction, possibly as a result of gravity.

Conclusions Gravitational forces may influence cell migration in vitro. This could be of significance in the planning of limbal transplantation, because a superior graft may be more likely to succeed than a gravitationally challenged inferior graft.

  • Cornea
  • gravity
  • limbal stem cell growth

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

The corneal limbus has been established as a repository for epithelial stem cells, which help in healing of corneal epithelial wounds and the maintenance of corneal epithelial integrity.1 The migration of cells from the limbus to the central cornea occurs in a centripetal manner. This centripetal migration is clearly evident in the healing of central corneal epithelial defects with an intact limbus.2 ,3 Re-epithelialisation of the central cornea is not established until the corneoscleral limbus is completely covered by epithelium.4 Attempts to correlate the proliferative potential of the limbus with the presence and distribution of the palisades of Vogt have proved inconclusive. Pellegrini et al5 have shown the nasal limbus to have the greatest proliferative potential. The limbal epithelial crypts, which have been shown to represent the niche for limbal epithelial stem cells, are randomly distributed along the limbus.1 Clinically, we have observed that following autolimbal or living related donor-derived allolimbal grafts (transplanted at the 12 and 6 o'clock positions) the epithelial cell sheet derived from the superior limbal graft extends downwards to cover up to the upper two-thirds of the corneal surface, whereas cells from the inferior limbal graft cover up to the inferior one-third of the cornea. The junction between the middle and inferior thirds of the cornea is therefore the meeting point of these two sheets. Furthermore, the Hudson–Stahli (HS) line, a horizontal line of epithelial iron deposition, is usually located at or near this junction.6–11 Studies using ultraviolet light, which unmasks the distribution of corneal ferritin, have confirmed that the HS line is a much more frequent finding than was previously thought.6 It is usually found to be higher on the nasal side of the cornea9 ,12–14 and can be observed in the young as well the old.15 The location of the HS line has been attributed to lid closure15 ,16 and to the inferior tear meniscus.17 ,18 However, another study has indicated that the line of lid closure and the HS line are at different levels.19 It is plausible that the HS line demarcates the junction between superior and inferior limbus-derived epithelial cell sheets. As the upper sheet is always larger than the lower by a third, we postulated that gravity may also influence corneal epithelial cell migration. This is relevant because in humans, the cornea is positioned vertically in the coronal plane for most of the waking hours. In contrast, when corneal epithelial cells are cultured in the laboratory the culture dishes are always placed in the horizontal plane. We tested this hypothesis in the laboratory by cultivating human corneal epithelial cells in the vertical plane.


Following corneal transplantation, 14 residual donor peripheral corneoscleral rims were used to culture corneal epithelial cells. All donor corneas were preserved and maintained in organ culture medium at 34°C in Eagles minimal essential medium containing 2% fetal bovine serum, penicillin 100 units/ml, streptomycin 0.1 mg/ml and amphotericin B 0.2 μg/ml for 3–4 weeks before surgery. The corneas were processed, stored, assessed and packaged according to Manchester Eye Bank quality standards.

The cell culture medium (CCM) that was used to feed the corneal epithelial cells consisted of Dulbecco's modified essential medium supplemented with 5% v/v heat-inactivated fetal bovine serum (Gibco Invitrogen, Paisley, Scotland, UK; cat no. 10270), insulin 4.5 μg/ml, 0.02 μg/ml gentamicin, 0.5 ng/ml amphotericin B (Gibco Invitrogen; cat no. 50-0640) and DMSO 0.5% v/v (Sigma-Aldrich, Munich, Germany; cat no. D2650) 100 μl cholera toxin (Vibrio cholerae, type Inaba 569B, Azide Free; EDM Chemicals, Merck KGaA, Darmstadt, Germany; cat no. 227036), 50 μl epidermal growth factor (recombinant human epidermal growth factor; cat no. 236-EG; R&D Systems, Minneapolis, Minnesota, USA).

The rims were placed in a petri dish (Sterilin, Stone, Staffordshire, UK; cat no. BS611) and a no. twenty-two blade (Swann-Morton, Sheffield, UK) was used to cut each rim into two equal halves under a dissecting microscope (Nikon Eclipse TS100; Nikon, Kingston upon Thames, Surrey, UK). The scleral portion of the corneoscleral rims was trimmed. Each half of the rim was then split horizontally to separate the epithelium from the endothelium, as described before.20 Briefly, the rim was divided through the middle of the stroma with a blade (dissecting curved, cat no. DC-M0052E; FEATHER, pfm medical, Cologne, Germany). The two sides of the split rim were grasped with a strong artery forceps and pulled apart, dividing the rim along its length into two strips: one containing the epithelium, basement membrane and stroma, and the other containing the rest of the stroma, Descemet's membrane and endothelium.

The endothelial strips were discarded, and the epithelial strips from each hemi-rim were placed in the petri dish with the epithelium facing upwards. Each half of the epithelial strip was further divided into five equal parts (explants), each of which was placed in the centre of a separate cell culture dish (2 mm grid from Nunc International, Rochester, New York, USA; cat no. 174926). The dish was marked along the 3–9 and 12–6 o'clock meridians and the long axis of each explant was orientated along the 3–9 o'clock meridian (figure 1). The explant was left on the culture dish, covered with a drop of medium and allowed to adhere to the plastic for 48 h. Thereafter the dish was filled with culture medium to submerge the explants. The CCM was changed every 2 days, taking care not to disturb the explants. All the explants were maintained at 37°C in the horizontal position for 7 days to ensure adherence to the plastic dish and cell growth from the explants.

Figure 1

The marked petri dish with the explant placed in the centre of the dish.

On day 8, cell growth was measured on the grid of the culture dish and explants with equal growth were paired. The chosen explants showed the same growth (between 1 mm and 2 mm; 1 mm = one square in the petri dish) towards the 12 and 6 o'clock meridians and equal number of squares (± one square) towards the 12 and 6 o'clock meridians in the grid. Squares that were covered by more than half the area were counted as one complete square and those that were less than half were not counted.

One dish containing an explant from each pair was placed vertically in a CCM-filled polypropylene container (80×50×50 mm) and the other matched explant of the pair was placed in a horizontal position in a similar polypropylene container (figure 2).

Figure 2

The polypropylene containers (pink) with the petri dishes (blue) inside. (A) Shows the vertically positioned petri dish. (B) Presents the horizontally positioned petri dish.

Each container was filled with 40 ml of CCM, enough to immerse the culture dishes completely and to maintain cell growth for 10–12 days. The size of the polypropylene container was selected deliberately so as to be able to fix the culture dishes in place either horizontally or vertically.

Thereafter, the culture dishes were transferred to the stage of a phase contrast microscope and the area and vertical length, both above and below the horizontal meridian, were determined by counting the squares in the grid.

Cell counts

Each explant was removed from the culture dish with a forceps. The remaining cell sheet was submerged under tryple E (Gibco Invitrogen, cat no. 12604-021) for 5 min to detach the cells from the culture dish. The more adherent cells were separated by using a cell scraper (Corning Incorporated Coster, Cambridge, UK; cat no. 3010). The cell suspension was transferred to a universal tube (Sterilin Ltd. UK, cat no. LS-M0570E) and the tryple E was neutralised by the addition of CCM. The tube was centrifuged for 5 min at 250 g and the cell pellet thus obtained was re-suspended in 500 μl of CCM and resuspended by gentle agitation of the universal tube containing the cells; 10 μl of cell suspension were extracted with a pipette and placed in an Eppendorf tube and 10 μl of trypan blue (cat no. T8154; Sigma, Poole, UK) was added. The cell suspension was used to charge the haemocytometer counting chamber and the cells counted using the 10× objective.

The number of healthy cells (unstained by trypan blue) in the area of 16 squares indicated by each of the five circles in the haemocytometer counting chamber were counted.

The haemocytometer is designed so that the number of cells in one set of 16 grid squares is equivalent to the number of cells ×104/ml.

The mean cell count was obtained by counting the cells in all five circles divided by five. This was multiplied by two to adjust for the 1:2 dilutions in trypan blue, and then by the volume to give the total cell count.

Statistical analysis

Data were analysed using a statistical software package, SPSS V.8.0. The normality of the data distribution was tested before analysis. The Student's t test was used for normally distributed data and the Wilcoxon signed rank test was used as a non-parametric alternative.


Epithelial cell growth was seen after 3–4 days of culture. Out of 14 donor rims used in this study, 10 were successfully cultured to completion of the experiment. The remaining four were excluded because of poor cell growth.

Two explants that showed between 1 mm and 2 mm of growth towards the 12 and 6 o'clock meridians (figure 3) were chosen. Before the explants were placed in either vertical or horizontal positions, the total number of squares showing cell growth was counted for each of the 20 explants. The mean number was 14.1±5.17, ranging from 8 to 23 squares per dish. Ten days after the explants were placed in the desired position, the mean number of squares containing cell growth was 54.7±14 squares (range 39–72) for horizontal explants and 54.9±13 squares (range 33–78) for vertical explants. In the vertical group, the mean number of squares showing cell growth towards the 6 o'clock position was 31.2±7 (range 20–41) and towards the 12 o'clock position was 24.6±6 (range 13–37). In the horizontal group, the means were 27.7±7 for the 6 o'clock position (range 19–39) and 27.0±7 for the 12 o'clock position (range 19–38; table 1). There was no significant difference between the total number of squares covered by cell growth between the culture dishes that were placed horizontally and vertically (p=0.9). In the culture dishes that were placed in the vertical position, the difference between the number of squares that had cell growth towards 12 o'clock and those growing towards 6 o'clock was statistically significant (p<0.001). The difference between the number of squares growing in the 12 and 6 o'clock positions was not significant for the horizontally placed explants (p=0.48).

Figure 3

Explant shows 1 mm growth (arrows) towards the 12 and 6 o'clock meridians after 8 days in the horizontal position. This was paired with another explant that showed the same growth before being placed in the required position either vertically or horizontally.

Table 1

Area of cell growth before and after changing orientation of culture plate to the vertical position

Mean cell counts for horizontal and vertical culture dishes were 7.18×104±2.4×104 cells and 7.26×104±2.2×104 cells, respectively. The difference between horizontally and vertically maintained cell growths was not statistically significant (p=0.7; table 2).

Table 2

Cell counts of cultured sheets


In total limbal stem cell deficiency with conjunctivalisation of the cornea, definitive management involves restoring the limbal function by transplanting autologous or allo-limbal tissue.21 It is usual protocol to transplant the donor explants in the 12 and 6 o'clock positions. After limbal stem cell transplantation, a common observation in the recipient eye is that outgrowth from the upper limbal stem cell graft tends to exceed that from the lower graft, with the cells derived from the upper explants covering the upper two-thirds of the cornea and that from the lower explants, the lower one-third. We have also observed (unpublished) that in some patients when cell growth from the superior limbal graft is delayed, cells from the lower 6 o'clock explant did not migrate upwards to cover the whole cornea. Rather, the cell sheet stopped progression upon covering approximately the lower third or half of the cornea. This difference could be due to gravity but other factors such as lid movements may also contribute. The effects of gravity, if any, on cell movement of the corneal epithelium has not been studied before.

The results of our experiments demonstrated that although there was no statistically significant difference in cell proliferation between the explants that were placed horizontally and vertically, the cells from the vertical explants migrated preferentially in a downward direction. This indicates that gravity could play a role in cell migration on the cornea, which in the normal anatomical position is oriented vertically in the coronal plane. Based on the clinical observations and this study it may be reasonable to assume that this pattern of cell migration reflects the normal physiological pattern of cell migration, with the upper limbus contributing more than the lower limbus. The meeting of the upper and lower sheets would thus form a natural junction at the line between the upper two-thirds and lower one-third of the corneal epithelial sheet. The boundary could be related to the HS line and also to the horizontal line of cornea verticillata (Farbys disease and amiodarone keratopathy), towards which the vortex lines converge.6

Corneal epithelial cells are intricately linked to the sub-basal plexus of corneal nerves and their anterior branches, which terminate within or in between the epithelial cells. In this context it is interesting to note that the orientation of the sub-basal nerve plexus is also predominantly vertical,22 wherein the upper vertical fibres meet with the lower vertically oriented fibres at the junction of the upper two-thirds and lower one-third. This upper–lower distribution of nerves is also obvious in the in-vivo confocal microscopy study of human corneal nerves reported by Patel and McGhee.23 Although they did not comment on this differential distribution, their images are similar to the whole mount histology images shown by Al-Aqaba et al,22 who also commented on the differential distribution.

The effect of gravity on the cell growth has been addressed previously.24–26 Rijken et al26 studied the effect of microgravity (which is a term used to describe weightlessness or zero gravity) on cells. They concluded that microgravity differentially modulates the epithelial growth factor and it also suppresses protein kinase C signalling, which affects the cell growth. Another study has shown that simulated microgravity inhibits wound healing and growth factor responses in the rat.27

Our study provides early laboratory evidence that gravity can favour corneal epithelial cell migration and expansion. Gravitational forces may influence cell migration in vivo, in the physiological maintenance of the corneal epithelial cell sheet and in the recipient eye after limbal stem cell transplantation. This could be of significance in the planning of limbal transplantation, because a superior graft may be more likely to succeed than a gravitationally challenged inferior graft. Moreover, the force of gravity could be used in stimulating ex-vivo stem cell expansion.



  • Competing interests None.

  • Patient consent Obtained.

  • Ethics approval Ethics approval for this study was obtained from the local ethics dommittee no 07/H0403/140.

  • Provenance and peer review Not commissioned; externally peer reviewed.