The unique organization of stromal collagen fibrils in the cornea is a particularly important ultrastructural feature that may influence the tissue's shape, biomechanical properties, and transparency. Corneal transparency, in particular, has been related to the arrangement of stromal collagen ever since the pioneering work of Maurice1 likened the cornea to a difiraction grating that causes the light scattered by collagen fibrils to destructively interfere in all directions except straight through the cornea. Subsequent calculations2,3 showed that conditions required for transparency could be satisfied by a lattice-like arrangement of collagen rather than a strict lattice in the physical sense. Observations from electron micrographs2 and data from experiments utilizing x-ray difiraction4 indicate that corneal collagen fibrils are arranged with short-range order extending over distances equivalent to a few fibril diameters. It is clear then, that since a loss of fibrillar order can lead to corneal opacification, some knowledge of the organization of collagen fibrils in rehydrated lyophilized cornea is desirable if this tissue is to be used for corneal surgery.
Present-day corneal x-ray difiraction experiments commonly utilize high-intensity x-rays from a synchrotron source, and thus enable us to measure the mean center-to-center spacing of regularly-arranged collagen fibrils quickly, while the specimens are still hydrated.5-10 We used such an experimental set-up to monitor the collagen interfibrillar spacing in lyophilized porcine corneas after rehydration.
MATERIALS AND METHODS
Ten porcine eyes were obtained from a local abattoir immediately after death and transported on ice to the laboratory. Within a few hours of harvest, 7.0-mm diameter corneal buttons were obtained and lyophilized. The tissue was stored at room temperature for several days prior to the experiments. Eight of the lyophilized corneas were rehydrated by immersing them in small amounts of balanced salt solution for various times ranging from 5 to 15 minutes, just before they were examined by synchrotron x-ray diffraction. Corneas were gently blotted to stop rehydration, weighed, and placed between two mylar windows in a specimen holder suitable for beamline X12B at the National Synchrotron Light Source, Brookhaven National Laboratory, USA. X-ray diffraction images were recorded, as described previously11, on a two-dimensional, position sensitive x-ray detector placed approximately 2.5m behind the specimen. Following a 60-second exposure, the specimens were weighed once more (they were typically a few percent lighter now) and the average of the weights before and after the exposure was used to calculate specimen hydration, quoted as its water content (wet weight -dry weight) divided by (wet weight) x 100% H2O).
Figure 1 : Micrographs of (A) a lyophilized porcine cornea, and (B) a lyophilized porcine cornea after it had been rehydrated so that 73.3% of its weight is water. It is clear that compacted collagen fibrils in the lyophilized cornea become more widely spaced when the tissue is rehydrated. Collagen stained with 1% phosphotungstic acid and 2% uranyl acetate (bar=130nm).
Figure 2: A cross-section of collagen in a lyophilized porcine cornea rehydrated so that 73.3% of its weight is water. In general, the packing appears normal, although some variation is evident. Collagen stained with 1% phosphotungstic acid and 2% uranyl acetate (bar=180nm).
When we analyzed the data, no angular dependence was evident on the x-ray patterns so they were circularly averaged around the beam center to produce radial diffraction profiles. Background radial profiles from the empty specimen holders were obtained similarly, scaled for differences in incident flux and then subtracted from the corneal profiles. Values for the collagen interfibrillar Bragg spacings from our eight rehydrated corneas were calculated from the profiles of the first-order equatorial x-ray reflections calibrated against the 67nm D-periodic x-ray reflections from moist rat tail tendon. We did not analyze untreated porcine corneas as part of this study. Values for the interfibrillar Bragg spacing of freshly thawed porcine corneas that contain between 74 and 78% water by weight are available in the literature.12,13 Collagen fibrils become closely packed when a cornea is frozen but the change is reversible and the fibrils regain their original arrangement when the tissue is thawed14; therefore, the data from freshly thawed tissue12,13 were taken to represent conditions close to physiologic.
The two remaining lyophilized corneas (one dry, one rehydrated) were examined by transmission electron microscopy as described previously.10
It is difficult to discern the characteristic fibrillar appearance of the stromal lamellae in lyophilized porcine corneas because the collagen is extremely close-packed and most fibrils are touching their neighbors (Fig 1A). With rehydration (73.3% water by weight), collagen fibrils become more widely spaced and a more normal collagen organization becomes evident (Fig 1B). Our cross-sectional electron microscopic images of the rehydrated specimen (Fig 2) are suggestive of subtle alterations in the normal lattice-like arrangement of collagen. Electron microscopic information of this type is, out of necessity, selective, and data regarding collagen spacings are better obtained via x-ray diffraction experiments, where the measurements represent an average for the whole thickness of the specimen through which the x-ray beam passes. These data for eight porcine corneas rehydrated to various levels are shown in Figure 3. The confidence limits represent the accuracy with which we could measure the position of the first order interfibrillar reflection, converted to real space. Since the lattice of corneal collagen fibrils swells in two dimensions when water is imbibed (the third dimension being along the fibril axis), we plotted our data as interfibrillar Bragg spacing squared against hydration.17 Figure 3 demonstrates that this quantity increases with hydration (Rp 2 = 0.844) although there is some variation evident in the data and some specimens with similar hydrations possess different interfibrillar Bragg spacings.
Figure 3: Graph of the collagen interi ibrillar Bragg spacing squared vs hydration for eight lyophilized porcine corneas rehydrated to different hydrations.
Previous synchrotron x-ray work has shown that freshly thawed porcine corneas (n=4) containing 76% (±2.0% water by weight) have a mean center-tocenter collagen interfibrillar Bragg spacing of 58.6nm (±4.5nm).12,13 The confidence limit of 4.5nm means that the upper and lower limits for the interfibrillar Bragg spacing squared in freshly thawed porcine corneas are 3981.61nmp 2 and 2926. 81nmp 2. When we insert these values into the straight line plot in Figure 3 (y=145.85; x=7912.5), we find that to attain an interfibrillar Bragg spacing for regularly arranged collagen that is the same as that for freshly thawed porcine corneas (74 to 78% water by weight), lyophilized porcine corneas need to be rehydrated so that they contain between 74.3 and 81.6% water by weight.
Subtle alterations in the arrangement of collagen in lyophilized corneas, similar to the electron micrographs of in vitro porcine tissue shown here, have been reported previously in human eyes grafted with lyophilized corneal tissue.15 However, no x-ray diffraction data exists regarding the organization of collagen in lyophilized corneal tissue.
If they are to possess values for the interfibrillar Bragg spacing that are in the same range as those from freshly thawed porcine corneas, lyophilized porcine corneas need to be rehydrated so that between 74.3 and 81.6% of their weight is water. This hydration range extends somewhat higher than that displayed by freshly thawed porcine corneas (74 to 78%)13, suggesting that, on average, lyophilized porcine corneas, when rehydrated, often require more water than is contained in freshly thawed porcine corneas if their regularly arranged collagen is to attain a similar spacing. The discrepancy between the upper limits of these ranges (78% water for freshly thawed corneas and 81.6% water for rehydrated lyophilized corneas) may seem fairly small when expressed as percentage of water, but, as Davson has pointed out16, it represents quite a significant increase in the amount of water per gram of dry material. We know this extra water in rehydrated lyophilized corneas is not distributing itself uniformly between the fibrils (this would lead to higher values for the interfibrillar Bragg spacing squared), and postulate that it is located in some intrastromal space(s) or compartment(s) that is either devoid of collagen or contains some randomly organized fibrils that, because of their nonperiodic arrangement, do not contribute to the interfibrfillar x-ray reflection. Such regions of the stroma, often termed lakes, have been invoked by numerous studies of the cornea.3,4,13,14,17-19 Moreover, it has been shown that lakes are more common in dried-then-rehydrated bovine corneas than in normally hydrated bovine corneas.4 This is in line with the present data, and implies that when corneas are dried, unlike when they are frozen14, we induce structural alterations in the collagen lattice that are not always reversible. This may be related to the fact that corneas exhibit a two-stage drying process17,20, whereby the initial water comes out of the extrafibrillar space, not from within the collagen fibrils themselves. Only later, when the cornea's water content drops below 50% by weight, is water also removed from within the fibrils. This causes a compaction of the collagen molecules that constitute the fibrils17,20 along with other, less well defined structural changes, both within the fibrils and on their surfaces.20
The variation in our x-ray data (Fig 3) shows that individual corneas that imbibe similar amounts of water do not necessarily attain similar collagen interfillar spacings. Presumably, this is due to an unequal distribution of water that will result in some corneas containing more lakes than others. However, the reasons for these inter-individual differences are not known. The concept of stromal lakes is fairly widespread.3,4,13,14,17-19 Theoretical considerations stipulate that if the dimensions of a stromal lake approach half the wavelength of light (> 200 to 250nm), it should scatter light3 and thus have a detrimental effect on corneal transparency. It is well known that swollen corneas become cloudy. We are not aware of any study that establishes the threshold at which lakes have a clinically observable effect. However, in some instances where lakes are known to be present, their effect is subclinical19, and even in a normally hydrated (bovine) cornea, it is thought that water does not occupy the extrafibrillar space uniformly.17 Nevertheless, as swollen corneas imbibe water into the extrafibrillar space17, we do not believe that it would be prudent to over-hydrate lyophilized corneal tissue prior to use.
The interfibrillar Bragg spacing measured in this study does not take into account the mode of packing of the stromal collagen fibrils. In their extensive x-ray study of the cornea, Worthington and Inouye21 analyzed the low-angle pattern in detail and concluded that stromal collagen fibrils are packed in a two-dimensional, liquid-like array, and that the actual mean center-to-center interfibrillar collagen spacing is best represented by multiplying the interfibrillar Bragg spacing by a factor 1.12. This is not the appropriate place to pursue the details of their interpretation- one that is now widely accepted5-7,9,10,12-14,16-19_but it means that the collagen interfibrillar spacing with packing taken into account should best be quoted as 65.6nm (±5.0nm).
On average, lyophilized porcine corneas are able to regain their normal interfibrillar collagen spacings when rehydrated so that between '74.3 and 81.6% of their weight is water. However, for reasons that are not known, some specimens seem to take up water differently than others that were handled in the same manner, and may possess more intrastromal lakes.
1. Maurice DM. The structure and transparency of the cornea. J Physiol (London) 1957;136:263-286.
2. Hart RW, Farrell RA. Light scattering in the cornea. J Opt Soc Am 1969;59:766-774.
3. Benedek GB. Theory of transparency of the eye. Appl Opti. 1971;10:459-473.
4. Sayers Z, Koch MHJ, Whitburn SB. Meek KM, Elliott GF, Harmsen A. Synchrotron x-ray diffraction study of corneal stroma. J Mol Biol 1982;160:593-607.
5. Quantock AJ, Meek KM, Ridgway AEA, Bron AJ, Thonar EJ-MA. Macular corneal dystrophy: reduction in both corneal thickness and collagen interfibrillar spacing. Curr Eye Res 1990;9:393-8.
6. Fullwood NJ, Tuft SJ, Malik NS, Meek KM, Ridgway AEA, Harrison RJ. Synchrotron x-ray diffraction studies of keratoconus corneal stroma. Invest Ophthalmol Vis Sci 1992;33:1166-1177.
7. Quantock AJ, Meek KM, Fullwood NJ, Zabel RW. Scheie's syndrome: the architecture of corneal collagen and distribution of corneal proteoglycans. Can J Ophthalmol 1993;28:266-72.
8. Rawe IM, Meek KM, Leonard DW, Takahashi T, Cintron C. Structure of corneal scar tissue. An x-ray diffraction study. Biophys J 1994;67:1743-1748.
9. Huang Y, Bron AJ, Meek KM, Vellodi A, McDonald B. Utrastructural study of the cornea in a bone-marrow transplanted Hurler syndrome patient. Exp Eye Res 1996;62:377-387.
10. Meek KM, Quantock AJ, Elliott GF, Ridgway AEA, Tullo AB, Bron AJ, Thonar EJMA. Macular corneal dystrophy: The macromoleculer structure of the stroma observed using electron microscopy and synchrotron x-ray diffraction. Exp Eye Res 1989;49:941-58.
11. Quantock AJ, Klintworth GK, Schanzlin DJ, Capel MC, Lenz ME, Thonar EJMA. Proteoglycans contain a 4.6A repeat in macular dystrophy corneas: x-ray diffraction evidence. Biophys J 1996;70:1966-72.
12. Gyi TJ, Meek KM, Elliott GF. Collagen interfibrillar distances in corneal stroma using synchrotron x-ray diffraction: a species study. Int J Biol Macromol 1988;10:265-269.
13. Meek KM, Leonard DW. Ultrastructure of the corneal stroma: a comparative study. Biophys J 1993;64:273-280.
14. Fullwood NJ, Meek KM. An ultrastructural, time-resolved study of freezing in the corneal stroma. J Mol Biol 1994;236:749-758.
15. Binder PS, Zavala EY, Baumgartner SD, Nayak SK. Combined morphologic effects of cryolathing and lyophilization on epikeratoplasty lenticules. Arch Ophthalmol 1986;104:671-679.
16. Davson H. Physiology of the Eye. 4th ed. London: Churchill Livingstone; 1980:96-97.
17. Meek KM, Fullwood NJ, Cooke PH, Elliott GF, Maurice DM, Quantock AJ, Wall RS, Worthington CR. Synchrotron x-ray diffraction studies of the cornea with implications for stromal hydration. Biophys J 1991;60:467-74.
18. Quantock AJ, Kratz-Owens KI, Leonard DW, Meek KM, Schanzlin DJ. Remodelling of the corneal stroma following lamellar keratoplasty: A synchrotron x-ray diffraction study. Cornea 1994;13:20-7.
19. Huang Y Tuft SJ, Meek KM. A histochemical and x-ray diffraction study of keratoconus epikeratoplasty. Report of two cases. Cornea 1996;15:320-328.
20. Fratzl P, Dexter A. Structural transformation of collagen fibrils in corneal stroma during drying. An x-ray scattering study. Biophys J 1993;64:1210-1214.
21. Worthington CR, Inouye H. X-ray diffraction study of the cornea. Int J Biol Macromol 1985;7:2-8.