Aqueous Humor Enhances the Proliferation of Rat Retinal Precursor Cells in Cultu - 《stemcells》干细胞学杂志干细胞之家



免疫细胞治疗专区	欢迎关注干细胞微信公众号

12 3 4 5 6 7 8 9 10 ... 23 下一页

返回列表

查看: 358545\|回复: 228	go Aqueous Humor Enhances the Proliferation of Rat Retinal Precursor Cells in Cultu [复制链接]

江边孤钓

版主

Rank: 7 Rank: 7 Rank: 7

积分: 0
威望: 0
包包: 483

楼主

发表于 2009-3-4 23:57 |只看该作者 |倒序浏览 |打印

作者：Jing Yanga,b, Henry Klassenc, Mette Priesa, Wei Wangb, Mogens H. Nissena作者单位：aLaboratory of Experimental Immunology, Department of Medical Anatomy, Panum Institute, University of Copenhagen, Copenhagen, Denmark;bDepartment of Ophthalmology, Third Hospital of Peking University, Peking University, Beijing, China; 0 V0 K" F3 \5 o* c6 ], K
   3 @* O( C: K: X
   ' W( z7 m0 e" j
      . @4 S6 l) J6 t
                     % ~6 z' ^) D) n* \7 P
            G" |1 t$ I9 U3 R0 \; ^3 i3 q: x
         6 d: P6 G- v6 v* l5 O
         8 @" R) i( x/ h' O3 J' y5 K4 @( J( _
         ) c* U2 L: }' V9 x; I  b) D0 |6 ^


      5 W" \+ y" _1 `- B
      ; I8 f& w" c- ?! Y- R. E% E" D
      【摘要】
   Aqueous humor has been shown to influence the proliferation of various ocular cell types, but the effect on immature retinal cells is not known. Here, the effect of pig aqueous humor on the proliferation of rat retinal precursor cells (RPCs) was investigated. RPCs were prepared from embryonic day 19 Sprague-Dawley rats and cultured in the presence or absence of aqueous humor from healthy pigs along with a medium consisting of Dulbecco's modified Eagle's medium:Ham's F-12 medium, N2 supplement, and epidermal growth factor. Proliferation was quantified by thymidine incorporation by as much as 317%, as compared with controls. Aqueous supplementation also increased both the number and size of RPC spherical aggregates ("spheres") over the first 4 days, consistent with increased proliferative activity. Using gel filtration and the in vitro proliferation assay, the growth-promoting activity of aqueous humor was localized to two different molecular mass ranges, namely, around 30 kDa and less than 1 kDa. Ascorbic acid was present in the lower molecular mass fraction, and use of this molecule reproduced some, but not all, of the proliferative activity present in aqueous humor. These results highlight the potential role of soluble factors present in the cellular microenvironment with respect to modulation of endogenous progenitor cell activity.
      【关键词】 Retinal precursor cells Aqueous humor Ascorbic acid Proliferation Low molecular mass protein
   INTRODUCTION
* N% {& G# a: i; W
Diseases of the retina are at present a major source of visual disability worldwide. Furthermore, the incidence of these conditions is increasing within the aging populations of the developed world. This situation reflects both the delicacy of the mammalian retina and its limited capacity for self-repair, along with a paucity of available treatments. The ongoing search for effective clinical options has been hampered by the complexity of the underlying biology, as well as by the relative specialization of the human retina for which no single animal model has provided an entirely adequate model.
0 A- p% f+ }7 J( T. N
One species that has been studied extensively with respect to retinal pathology is the rat, despite a number of evident dissimilarities, including the absence of a specialized fovea. Notable among the rat models of retinal degeneration is the Royal College of Surgeons (RCS) rat .! D9 s9 {# F# }, R: j, E4 r+ y! l; i
8 k; \! G' h8 O3 v, y) h, n3 E) L+ @
Initial work involving the transplantation of adult hippocampal progenitor cells (AHPCs) to the rat retina showed widespread integration of donor cells after grafting to neonatal recipients but not in the mature retina of normal syngeneic adults , the need for substantial production of rod photoreceptors has stimulated the search for similar cells from within the neural retina.+ j- j1 r& c. H$ q& n6 `! v

In the mouse, retinal precursor cells (RPCs) have been isolated and propagated for extended periods. These cells passage continuously for over a year, express the retinal markers recoverin, rhodopsin, and cone opsin following subretinal transplantation, and are associated with rescue of host photoreceptors and improved light sensitivity compared with control animals . Although the reason for this lack of sustained proliferation of rat RPCs is not clear, it is possible that essential factors normally present during development are not supplied under the standard culture conditions used.9 w3 y# L6 j+ Q3 `: f3 U
2 O7 \# g6 z, R" h2 l  h6 B, {5 A
One way to explore the above possibility is to modify culture conditions in an attempt to better reflect the native microenvironment of the developing retina. Among the most accessible strategies of this type would be to supplement cultures with aqueous humor. In the adult eye, fluid secreted from the ciliary body not only gives rise to the aqueous humor in the anterior segment but also saturates the vitreous body and passes through the intercellular compartment of the neural retina as it leaves the posterior segment, partly through the action of cells of the RPE layer . Although size constraints make the composition and physiology of the extracellular fluid present in the embryonic rat retina difficult to determine, aqueous humor is readily obtainable from the eyes of large domesticated mammals. Given the importance of the rat as an experimental model of retinal pathology, the current study was directed toward exploring the potential effects of adult porcine aqueous humor on the proliferation and survival of rat RPCs in culture.$ V7 E4 T5 j$ o4 L% K

MATERIALS AND METHODS7 @5 f) ^  S" k

Donor Animals7 j. ~) [( v5 e8 O4 I: G
1 b8 m# O; \3 V* e: T) Q* s- J
Timed-pregnant Sprague-Dawley rats (Charles River, Sulzfeld, Germany, http://www.criver.com) were used to obtain embryonic day 19 (E19) embryos for isolation of precursor cells. These animals were handled according to the Association for Research in Vision and Ophthalmology (ARVO) Statement for the Use of Animals in Ophthalmic and Vision Research. Porcine eyes for obtaining aqueous samples were purchased from a local abattoir (Denmark).: K1 n& |  Z4 P% x* N& p

Cell Isolation and Culture

E19 embryos were harvested from timed-pregnant Sprague-Dawley rats. The eyes were dissected from the embryos with care taken to minimize the amount of adherent mesenchymal tissue. Enucleated eyes were collected in Hanks' balanced salt solution (HBSS) and then transferred to fresh HBSS in a separate culture dish. The optic nerve and remaining mesenchymal tissue were removed before isolating the retina. This approach was taken to avoid possible contamination of retinal isolates with brain-derived cells. The retina was carefully teased away from the RPE, and the central portion of the retina and the optic nerve head was removed and discarded. The isolated neural retina was collected in a dish, dissociated into small pieces, digested in 0.05% trypsin (Invitrogen, Carlsbad, CA, http://www.invitrogen.com) for 5 minutes at 37¡ãC, and then gently triturated using a 1-ml fire-polished glass Pasteur pipette to release single cells. Samples were then passed through a 40-µm filter. The resulting cell suspension was centrifuged at 300g for 6 minutes, and the pellet was resuspended in fresh serum-free culture medium composed of Dulbecco's modified Eagle's medium/Ham's F-12 medium with 1 mM glutamine (Gibco, Copenhagen, Denmark, http://www.invitrogen.com), N2 supplement (1%; Invitrogen), and 20 ng/ml epidermal growth factor (EGF) (Invitrogen). Eosin red was used to evaluate viability, and cell density was adjusted to 0.4 x 106 cells per milliliter for plating. Cultures were incubated at 37¡ãC under 5% CO2.4 o& `% \* K9 j
5 v; y; [, K7 E/ b* ]9 q
Every 3 days, the medium was changed completely, and after 6¨C8 days in culture primary cellular aggregates were collected and incubated in cell dissociation buffer (Gibco) for 5 minutes at 37¡ãC. These cellular aggregates characteristically displayed a roughly spherical shape, analogous to those cellular clusters elsewhere described as "neurospheres." To avoid potential overinterpretation, they are referred to here simply as "spheres." These spheres were then gently triturated using a fire-polished Pasteur pipette to release single cells, which were then plated at the same density as before and cultured in the identical manner. To establish the extent and efficiency with which the sphere-forming cells retained their ability to form secondary (first passage ) spheres, their numbers were counted before plating.* L8 n, }  m; U
7 c1 g4 g/ i, v  Y0 H0 ?
Preparation of Aqueous Humor

Samples of aqueous humor were obtained from enucleated pig eyes by anterior chamber paracentesis within approximately 1 hour of death. The eyes were transported on ice and kept at 4¡ãC until use. The samples were collected using a 25-gauge needle with great care taken to ensure that the needle did not touch the iris, lens, or corneal endothelium during collection. Samples were centrifuged for 5 minutes at 700g, and supernatants were collected. These were either used in cell culture, stored at ¨C80¡ãC, or fractionated by gel filtration chromatography (GFC).3 z! F( ?7 P0 e$ e, r( q% l

Proliferation Assay

Fresh retinal isolates were plated into 96-well flat-bottomed plates at 0.4 x 106 cells per milliliter as a single-cell suspension in the experimental media of various compositions being tested. To these plates was added thymidine incorporation was determined using a Microplate Scintillation and Luminescence Counter (PerkinElmer, Wellesley, MA, http://www.perkinelmer.com) to generate a proliferation curve for the time period from day 0 to day 4.# ^+ F  z/ a2 {$ t4 N. |6 g
9 A0 v6 I9 O2 L. v: j+ ~, c: J
Morphological Analysis

Images of the primary cell isolates and subsequent cultures were captured using a Leica DM IRB microscope equipped with a Leica DC 300 camera and computer running IM 50 image analysis software (Leica, Heerbrugg, Switzerland, http://www.leica.com). Images of the central field of RPC cultures were recorded daily at a magnification of x50. Measuring and counting were performed using a Nikon 600 Imagepro cell statistics program (Nikon, Tokyo, http://www.nikon.com). Spheres were counted, and cross-sectional areas were measured. Many "spheres" were actually oblong in shape; hence, diameter frequently varied within a given sphere. For this reason, cross-sectional area was considered a more accurate assessment of sphere size than diameter. The threshold parameters for a sphere to be counted were defined as cross-sectional area greater than 800 µm2 and minimum diameter greater than 30 µm.: w5 n# S" I: I/ s

Statistical Analysis% }. ^" `6 n. c  p3 b

Analysis of variance was performed to statistically evaluate differences in the proliferation of cells grown in media of different compositions.
6 f# j8 K1 {0 x4 [; R7 r
Flow Cytometry# l; ?  E0 O& ~9 k5 e$ v0 c

Either primary or cultured rat RPCs were pelleted by centrifugation (500g; 5 minutes; room temperature) and resuspended as a single-cell suspension of approximately 106 cells in 100 µl of phosphate-buffered saline (PBS) (without serum) using a fire-polished glass Pasteur pipette. Cell membranes were permeabilized by saponin for 15 minutes prior to labeling, followed by a PBS wash. Cells were then incubated with 1 µg of anti-nestin primary antibody, or isotype control, for 30 minutes on ice. Unbound antibody was removed with two washes in PBS of 5 minutes each. Cells were then incubated with 1 µg/106 cells of the secondary antibody PE for 30 minutes on ice. Washes were again performed as described previously. Cells were then fixed in 1% paraformaldehyde. Automated analysis and sorting were performed using a FACS Caliber (Becton, Dickinson and Company, Franklin Lakes, NJ, http://www.bd.com).
2 j- t7 a% \/ a# e
Gel Filtration Chromatography

These experiments used an AKTA Explorer system (Amersham Biosciences, Hillerod, Denmark, http://www.amershamhealth.com). Separations were performed using a Superose 12HR 10/30 gel filtration column (Amersham Biosciences) with optimal separation range between 1,000 and 300,000 Da. Potassium phosphate buffer (50 nM, pH 7.4), containing 100 mM KCl and 0.6 mM sodium azide, was used in all separations with a constant flow rate of 0.5 ml/minute. This buffer was also used for reconstitution and dilution of protein standards. The volume of aqueous humor applied to the column was 250 µl. A gel filtration calibration kit for molecular mass proteins was purchased from Sigma and included blue dextran (2,000 kDa), thyroglobulin (669 kDa), apoferritin (443 kDa), ß-amylase (200 kDa), alcohol dehydrogenase (150 kDa), albumin (66 kDa), carbonic anhydrase (29 kDa), cytochrome C (12.4 kDa), and cytidin (243 Da). The elution volume of each standard protein was determined, and the logarithms of their molecular masses were plotted against elution time. This calibration curve was used to determine the approximate molecular mass of the protein peaks seen in the chromatograms.
1 k0 O4 g- `' c; B; J
Evaluation of the Proliferative Activity of Aqueous Fractions

Primary rat retinal cells in a volume of 160 µl and at a density of 0.15 x 106 cells per milliliter were plated into individual wells of flat-bottomed 96-well plates containing 40 µl (20% by volume) of a specific aqueous humor fraction, previously separated from fresh aqueous humor by GFC. Three wells were used for each fraction. In addition, control wells containing 160 µl of cells and 40 µl (20%) of either unfractionated aqueous humor or PBS were also plated at this time. Cellular proliferation was assayed by thymidine incorporation on day 2, as described previously.6 Z, O6 q5 e" K
' I. M( [' N! q9 {" d3 D. f
Spectrophotometric Analysis of Aqueous Humor
+ _/ M4 p) x0 z4 m
To examine the molecular composition of aqueous humor, specifically with respect to the presence of ascorbic acid, UV absorption spectra were obtained on three separate samples, and the data were compared by using a UV spectrophotometer (Beckman DU 650 spectrophotometer; Beckman Coulter, Fullerton, CA, http://www.beckmancoulter.com). These included unfractionated porcine aqueous humor (fresh, diluted 10-fold in PBS), the maximum aqueous fraction exhibiting peak absorbance for UV light of 280 nm (corresponding to an elution volume of 18 ml), as well as pure ascorbic acid (Merck & Co., Whitehouse Station, NJ, http://www.merck.com; as a 0.1 mg/ml solution in PBS). The length of the light path in the quartz cells used was 1 cm.

Evaluation of the Proliferative Effect of Ascorbic Acid on Rat RPCs
* S( M3 P% m9 F7 J: `; B7 j1 J
Primary rat retinal cells in a volume of 160 µl of standard EGF-containing growth media were plated into individual wells of flat-bottomed 96-well plates at a density of 0.24 x 106 cells per milliliter. To each well was also added 40 µl (20% by volume) of either ascorbic acid (0.1 mg/ml in PBS), porcine aqueous humor, or additional standard EGF-containing growth media. Three wells were used for each sample type. Cellular proliferation was assayed by thymidine incorporation from day 1 to day 3, as described previously. In an additional experiment, cells were plated as before and the proliferative effects of ascorbic acid were assayed, in this case over a range of concentrations (each 20% supplement contained from 0.03125 to 2 mg/ml ascorbic acid in 40 µl of PBS; hence, the total concentration in each well varied from 0.00625 to 0.4 mg/ml).

RESULTS
3 r8 E# N* A, }9 E" h3 z$ e+ W
Morphological Characteristics of Rat RPCs Cultured in Serum-Free Medium
! f0 C  x+ |3 |+ {; v+ D
Primary E19 retinal dissociates showed evidence of cell division on day 1 after plating, forming small spherical aggregates in serum-free media and uncoated flasks. After 1 week in culture, the spheres had increased in size while decreasing in number, consistent with continued proliferation together with the ongoing coalescence of these highly adherent cellular clusters observed in culture. At time points longer than 10 days, the spheres appeared more loosely organized, suggestive of decreased viability. Eosin staining confirmed that the spheres contained considerable numbers of nonviable cells at this last time point, particularly within their interior. After P1, proliferating cells typically exhibited limited cytoplasm and few, if any, cellular processes. It was not uncommon to observe aggregates of small, phase-bright cells with no processes. In contrast, a few profiles exhibited a larger volume of cytoplasm, multiple cellular processes, and an irregular shape consistent with more differentiated cells.

Aqueous Humor Enhances Proliferation of Rat RPCs in Culture

Image analysis of rat E19 retinal precursors cultured with and without 20% aqueous supplementation showed evidence of increased sphere formation in the aqueous-supplemented medium (Fig. 1). Beginning on day 0, and continuing to day 4, it was apparent that more spheres reached threshold criteria in the aqueous-supplemented medium at each time point and that these spheres were larger in size. This impression was confirmed by quantitative analysis, with both number (Fig. 2A) and size (Fig. 2B) of spheres being significantly higher in the aqueous-supplemented cultures, as compared with growth factor-containing medium alone.9 Q! f( b# o* r! ]4 E
8 V+ h; }' u: e  E. t# c" k
Figure 1. Photomicrographs of rat RPCs in epidermal growth factor-containing culture medium (A¨CE) or the same medium supplemented with 20% aqueous humor (F¨CJ). Data are shown sequentially from day 0 (top row) to day 4 (bottom row). Cells grown in the presence of 20% aqueous humor appeared to form larger spherical cellular aggregates (spheres) over the course of this time period. Magnification, x50. Scale bar = 250 µm.
" O) V" X/ @+ ~4 t5 a6 y
Figure 2. Quantitative analysis of sphere formation in culture. (A): Number of spherical cellular aggregates, or spheres, with cross-sectional area >800 µm2 is shown for each time point. Spheres of this size increased in number up to day 3, then decreased on day 4 for both treatment conditions. Greater numbers of spheres were found in aqueous-supplemented medium at all time points beyond the day of initial plating (day 0). (B): Size of spherical cellular aggregates with cross-sectional area >800 µm2 is shown for each time point. The aqueous treatment condition was associated with significantly larger spheres on days 3 and 4. Data represent the means of three samples from same plating ¡À SD. *, p 4 G* A4 }3 }( H- m8 _5 v

Further examination of the images shown in Figure 1 revealed a progressive decrease in the density of subthreshold profiles, particularly in cultures with many suprathreshold spheres. The major factor likely responsible for this finding is adherence of growing spheres to each other. This phenomenon was directly observed in these cultures and can be seen on close inspection as the multilobulated profiles present at later time points (Fig. 1 ). Coalescence of cultured spheres has been observed previously with a variety of other CNS precursor cell types (H. Klassen, unpublished observations). Thus, the observed increases in number and size of spheres may have contributions from both proliferative and adherent mechanisms, thereby illustrating one of the limitations inherent in using sphere formation alone as an assay for proliferation.
- p& T! d. Z6 Y
Having obtained morphological evidence suggesting that 20% aqueous supplementation promotes the proliferation of rat RPCs beyond the level seen with growth factor-containing medium alone, we sought to re-examine this finding using a quantitative measure of cellular proliferation. Thymidine incorporation assays confirmed increasing proliferation of E19 rat retinal precursor cells in medium containing the growth factor EGF up to days 2¨C3 in culture, followed by a subsequent decline (Fig. 3). Twenty percent aqueous supplementation resulted in significantly higher proliferation at all time points; however, the overall growth pattern again showed slowing of proliferation from days 3 to 4 (Fig. 3). Controls in which EGF-containing growth medium was supplemented with PBS showed slightly less benefit than growth medium alone, perhaps resulting from a dilutional effect on EGF concentration (Fig. 3A). As expected, the use of medium lacking the mitogenic growth factor EGF resulted in an immediate decline in proliferative activity, dropping to undetectable by day 3 (Fig. 3A). Supplementation of mitogen-free medium with 20% aqueous humor helped to sustain proliferative activity, albeit at levels less than that seen when aqueous was combined with EGF (Fig. 3A). Culturing rat RPCs in 100% aqueous humor resulted in an immediate decline in proliferative activity, dropping to negligible levels by day 3 (Fig. 3B).

Figure 3. Proliferation of retinal precursor cells (RPCs) in different culture media. (A): Thymidine uptake by RPCs in medium with or without EGF or AH, from day 1 to day 3. Rat RPCs grown in EGF-containing medium with 20% aqueous supplementation showed evidence of increased thymidine incorporation at all time points, as compared with either EGF-containing medium alone or EGF-containing medium supplemented with 20% PBS as a control. In the absence of either EGF or aqueous, thymidine incorporation rapidly declined to negligible levels. Supplementation of EGF-free medium with 20% aqueous significantly improved thymidine incorporation. (B): Dose-response relationship for AH and rat RPC proliferation. Supplementation of EGF-containing medium with 10% aqueous increased thymidine uptake at all time points measured, as did 20% aqueous to an even greater extent. Fifty percent aqueous supplementation was also associated with increased thymidine uptake, but not to the levels seen with 20% aqueous. One hundred percent aqueous was not associated with substantial thymidine uptake. Data represent the means of three samples from same plating ¡À SD. *, p % V. L1 H; K2 a
  {; q# O0 |! Q% Z' I( w% U
The dose-response relationship between the amount of aqueous humor added and the level of thymidine incorporation, as noted above.# B) e5 Y9 s; R

Comparison of Activity in Fresh and Freeze-Thawed Aqueous Samples

The protein content of freeze-thawed aqueous humor could potentially be altered through the presence of endogenous proteolytic activity. To rule out this possibility, proliferation results for rat RPCs obtained using fresh aqueous humor were compared with those obtained using previously frozen aqueous humor (stored at ¨C80¡ãC for less than 1 month). No difference was found between these conditions, whereas a decrease in stimulation by frozen-thawed aqueous humor would be expected if proteolytic activity were present in these samples.2 h2 J3 I" g* _, j  K6 P$ v4 z* ~

Stability of Marker Expression with Exposure to Aqueous Humor1 M9 [+ p+ W( S, l- w# W3 |0 g) ^

Flow cytometry was used to evaluate the expression of phenotypic markers. The percentage of cells positive for the primitive neuroepithelial marker nestin was 23% in primary E19 retinal isolates (Fig. 4B). Markers associated with more mature retinal cell types were not evident. In particular, the RGC-associated marker Thy-1 (CD90), the astroglial marker GFAP, the bipolar marker PKC, and the amacrine marker syntaxin were not detected using this method (data not shown). After 4 days in culture in the presence of 20% aqueous supplementation, the percentage of nestin-expressing cells was again 23% (Fig. 4D), and the other markers were again not detected. Taken together, these data are most consistent with stable neurodevelopmental marker expression and give no indications that aqueous humor induces RPCs to differentiate, although this latter possibility cannot be ruled out.2 R+ [7 K9 S& h8 Y

Figure 4. Nestin expression as a function of exposure to aqueous humor. Flow cytometry was used to evaluate expression of the intracellular marker nestin by rat retinal precursor cells (RPCs). (A) and (B) show samples of primary rat RPC isolates, with antibody-labeled cells showing positive nestin expression (B) compared with isotype controls (A). After 4 days of culture in the presence of 20% aqueous humor (C, D), rat RPCs showed continued expression of nestin (D) at a level similar to that seen in primary isolates, again compared with isotype control (C). Abbreviation: SSC, side scatter.
4 V1 ]2 U- m2 e  A! V+ J
Specific Aqueous Fractions Enhance Proliferation of Rat RPCs$ m) h2 F4 _1 {( a# \  L  l0 O0 x
% O! `% h# ?1 D: x
To identify components responsible for growth-promoting activity, porcine aqueous humor was fractionated by GFC. The separation process was monitored by absorbance at both 215 nm and 280 nm for the detection of peptide bonds and aromatic residues. Subsequently, the individual fractions were tested for their ability to promote the proliferation of freshly isolated RPCs using a thymidine incorporation assay. As shown, two different molecular mass entities with growth-promoting activities were identified (Fig. 5). One of these had a molecular mass around 30 kDa, and the other had a molecular mass less than 1 kDa.

Figure 5. Characterization of the growth-promoting activity present in aqueous humor by use of size exclusion gel chromatography. Gel filtration of 250 µl of pig aqueous humor was performed on a Superose 12HR 10/30 column using a flow rate of 0.5 ml/minute. The separation and elution was done in a physiological phosphate-buffered saline buffer, pH 7.4, at room temperature. Fractions of 1 ml were collected for testing of growth-promoting activity. Each of the fractions obtained by gel chromatography were assayed via thymidine incorporation assay for growth-promoting activity in triplicate wells containing rat retinal precursor cells. Absorbance was measured at 215 nm and 280 nm wavelengths. Proliferation shown as cpm. Arrows indicate molecular mass markers (as described in Materials and Methods). Abbreviation: AU, absorbance unit.

The Low Molecular Mass Aqueous Fraction with Proliferative Activity Contains Ascorbic Acid$ q% ~' K( v  T) q

A small molecule known to be concentrated in aqueous humor is ascorbic acid. The presence of ascorbic acid in a solution is characterized spectrophotometrically by a distinct absorption band with a maximum at 265 nm. To assay for the presence of this molecule in the proproliferative low molecular mass fraction of aqueous humor identified above, UV absorption spectra were used to compare this fraction with a 0.1 mg/ml solution of ascorbic acid as well as with unfractionated aqueous humor. Results confirmed a peak with an absorbance maximum at 265 nm in both the ascorbic acid standard and in unfractionated aqueous, together with a prominent peak at this wavelength in the proproliferative low molecular mass aqueous fraction (Fig. 6). These data show that ascorbic acid is present in this fraction. In gel chromatography, ascorbic acid was found to elute at a position identical to that of the low molecular mass proproliferative fraction (data not shown).
) G9 Z- e9 G# E! H- _' m9 K9 S  D
Figure 6. UV absorption spectra of aqueous samples and ascorbic acid. Whole porcine aqueous humor (solid line) was compared with the low molecular mass aqueous fraction with demonstrated proproliferative activity (dashed line) and with 0.1 mg/ml ascorbic acid solution (dotted line). A maximum was seen in all curves at a wavelength of approximately 265 nm, consistent with the presence of ascorbic acid in each specimen. Length of light path = 1 cm.

L-Ascorbic Acid Enhances the Proliferation of Rat RPCs

To investigate the potential role of ascorbic acid in the proliferative activity of aqueous humor, this low molecular mass compound was added to rat RPC cultures. On day 1, wells with ascorbic acid (20% supplementation with a solution of 0.1 mg/ml, hence 0.02 mg/ml final concentration) exhibited increased thymidine uptake above the level found in wells containing standard EGF-containing medium and comparable to the level obtained with of 20% aqueous supplementation (Fig. 7A). The proproliferative effect attributable to ascorbic acid was no longer apparent on days 2 and 3, consistent with rapid loss of this molecule in culture. In contrast, the advantage of unfractionated aqueous persisted over this time period. These results are consistent with the concept that ascorbic acid accounts for some, but not all, of the proliferative activity present in aqueous humor.7 ~4 q  A* P; M' V% X  }
! e1 ]. ~  p: l8 F/ t: g0 T
Figure 7. Proliferation of rat retinal precursor cells (RPCs) plated at low density in medium containing L-AA. (A): Comparison of standard EGF-containing growth medium versus the same supplemented with either 20% porcine aqueous humor or 20% L-AA as a 0.1 mg/ml solution in phosphate-buffered saline (PBS) (0.02 mg/ml total concentration). AA enhanced proliferative activity on day 1 to a level similar to that of aqueous humor. On days 2 and 3, AA showed no benefit over EGF-containing medium alone, whereas aqueous continued to do so. (B): Dose-response relationship between AA and rat RPC proliferation over a range of concentrations. The dosages shown refer to the concentration of AA in PBS, in each case delivered as a 20% (by volume) supplement to RPCs in standard EGF-containing medium. The peak activity seen at the 0.125 mg/ml dosage corresponds closely to the level of AA provided by 20% aqueous supplementation (described in Results), whereas higher levels of AA are progressively less beneficial. To highlight any proproliferative activity, RPCs were plated at 0.24 x 106 cells per milliliter throughout. Abbreviations: AA, ascorbic acid; AH, aqueous humor; EGF, epidermal growth factor.
0 ]# J1 m6 d9 p2 w2 H7 u
To further investigate this proliferative effect, ascorbic acid was added to rat RPC cultures over a range of concentrations. These experiments revealed a dose-response curve, peaking at the 0.125 mg/ml value (as 20% supplementation, thus corresponding to 0.025 mg/ml total ascorbic acid), and progressively decreasing at both higher and lower levels (Fig. 7B). For comparison, the level of ascorbic acid in whole porcine aqueous humor is 0.1 mg/ml, corresponding to the 0.5 mg/ml value of 20% supplementation shown in Figure 7B. The level of ascorbic acid in medium supplemented with 20% ascorbic acid is therefore 0.02 mg/ml, corresponding to the value of 0.1 mg/ml in Figure 7B. This value falls between the two highest performing doses of 0.125 and 0.0625 mg/ml and is thus consistent with the proproliferative effects of 20% aqueous supplementation noted earlier in this study.

Of note, in the experiments shown in Figure 7A and 7B, cells were plated at lower densities than is optimal for proliferation (0.24 x 106 cells per milliliter, as opposed to 0.4 x 106 cells per milliliter) in an effort to better delineate any proproliferative activity present. This accounts for the progressive decline in proliferative activity over time observed in all samples, versus that seen when cells were cultures at higher density (Fig. 3).
4 l; C' E2 L6 G5 ]; E5 b2 R
DISCUSSION
' h3 {, m4 \+ H
The results of this study show that supplementation of growth medium with aqueous humor enhances the proliferation of rat retinal precursor cells in culture. Analysis of the molecular composition of porcine aqueous by GFC revealed that two major peaks, one around 30 kDa and the other less than 1 kDa, retained a proproliferative influence on rat RPCs. Further analysis of the low molecular mass fraction showed that it contained ascorbic acid, a small molecule known to be concentrated in aqueous humor. Additional experiments demonstrated that ascorbic acid is itself proproliferative for primary rat RPCs and reproduces some, but not all, of the activity present in aqueous humor. Although the molecular basis of the larger molecular mass fraction remains to be determined, the role of cytokines in cellular proliferation and differentiation has been well established. It is therefore of interest that the 30-kDa peak is at least consistent with a molecule of this type.
# p; G: t2 B% F) w1 g
Aqueous humor has been reported to stimulate the proliferation of a variety of cell types in culture, including those derived from the Tenon capsule, lens epithelium, corneal endothelium, and iris epithelium  aqueous samples were tested for growth effects on cultured cells or explants from various tissue sources. In general, the collective observations of these studies indicate that in the presence of high concentrations of aqueous humor, there is significantly less growth than in parallel aqueous-free control cultures. Since the opposite conclusion was drawn from our experiments, it is important to consider whether different concentrations of aqueous humor have different affects on the cells' proliferation.

In the present study, supplementation of proliferation medium with 20% aqueous humor induced a statistically significant increase in thymidine uptake by rat RPCs over that seen with standard proliferation medium alone. Furthermore, 10% and 50% aqueous humor also appeared to promote proliferation. Aqueous humor appeared to support, to some extent, the proliferation of rat RPCs in the absence of any added growth factors, although proliferation and survival were poor under these conditions. In contrast, growth in 100% aqueous humor led to decreased proliferation of rat RPCs as compared with using proliferation medium alone. Thus aqueous humor cannot entirely substitute for proliferation medium; however, the combination of aqueous and proliferation medium appears to be superior to proliferation medium alone. This suggests that aqueous humor contains a proproliferative activity that acts synergistically with EGF-containing proliferation medium.
: N0 d" V0 o! ]4 j. _+ e  t
The aqueous humor supplies nutrients to avascular structures in the anterior eye, including the cornea, lens, and trabecular meshwork, as well as contributing to the bulk flow of fluid through the vitreous compartment. In terms of composition, albumin (0.596 µg/µl) is the major protein component in human aqueous humor , and lysozyme (0.064 µg/µl) but not IgM. Other compounds include myocilin, transferrin, basic fibroblast growth factor (bFGF), cystatin, superoxide dimutase, glutathione peroxidase, glutathione reductase, lactic acid, latent collagenase, tumor necrosis factor-ß, -tocopherol, glutathione, and ascorbic acid.
) N( k$ T6 ~) }  v4 v0 V5 m! v
Particularly well studied is the composition of primate aqueous humor compared with serum . In aqueous humor, the concentrations of phosphates, calcium, urea nitrogen, creatinine, glucose, cholesterol, glutamic-pyruvic transaminase, glutamic-oxaloacetic transminase, lactic dehydrogenase, creatinine phosphokinase, and alkaline phosphatase were lower than in serum. Conversely, aqueous humor contained higher concentrations of chloride, bicarbonate, uric acid, lactate, and ascorbic acid than did serum. The concentrations of sodium, potassium, and magnesium were nearly equal in the two fluids. Despite a several-hundred-fold lower protein concentration, aqueous humor osmolality was equal to serum osmolality. Aqueous humor was deficient in a number of amino acids compared with serum. Two amino acids¡ªcysteine and valine¡ªwere of equal concentration in both fluids. Serine, methionine, leucine, phenylalanine, and tyrosine appeared to be more concentrated in the pooled aqueous humor than in the corresponding serum.! m, i% N; T' K, a, l& Y, n, i: i1 A4 |

Aqueous composition has also been evaluated for the presence of immunomodulatory molecules. Various ocular microenvironments have long been recognized as sites of relative immune privilege, and immunosuppressive neuropeptides found in aqueous humor are central to this immunoregulation . These neuropeptides include -melanocyte-stimulating hormone, vasoactive intestinal peptide, calcitonin gene-related peptide, and somatostatin. Along with transforming growth factor-ß2, the neuropeptides target specific cells and pathways in innate and adaptive immunity. These aqueous humor factors prevent pathogen-induced inflammation and activation of Th1 cells, while promoting induction of regulatory T-cells. Therefore, the ocular microenvironment, through the constitutive production of immunosuppressive factors found in aqueous humor, maintains immune privilege by manipulating regional innate and adaptive immunity away from inflammatory responses. These factors could also potentially influence the proliferation of stem or precursor cells.  }6 p2 A# c) J9 ~/ U# S/ r! G  I
" _4 s6 ~9 \. n
The separated polypeptide components of aqueous humor appear to have differential effects on the proliferation of RPCs. Although aqueous fractions did not generally stimulate as well as unfractionated aqueous, this could relate to dilutional effects resulting from the fractionation process. With regard to potential species-related effects, in this study porcine aqueous humor was used to stimulate the proliferation of rat RPCs. Although these two species differ greatly from one another, many previous studies have indicated that cytokines frequently produce effects across species lines. For instance, recombinant human growth factors can be used to culture rodent CNS progenitor cells. Indeed, the porcine aqueous used here was seen to have a growth-promoting effect on the rat cells in this study. This is not to say that porcine aqueous humor might not have an even greater proliferative effect on porcine RPCs, although this possibility would need to be examined.- ^% ~) W. a9 I' A
5 z9 @5 Z; Z  X, K; X) q& |
Ascorbic acid, as opposed to peptides, exhibits no such species differences. This molecule, widely recognized for its role as an antioxidant, is not routinely included in standard proliferation protocols used for culturing progenitor cells from the developing central nervous system. Although the role of ascorbic acid in the eye might seem reasonable based on the concentration of this molecule in intraocular fluid, there is evidence that it can also play a role in the ontogenetic status of precursor cells in other systems, notably, enhancing the differentiation of a variety of more primitive cells into osteoclasts . Whether this disparity in results as compared with the present study reflects a biological difference between retinal and forebrain precursors as opposed to specific culture conditions remains to be elucidated.4 T# F: ~# I% S3 C" Q8 @7 w5 U
7 L* X- R) L+ `+ S& g' z3 n
Conclusions. S& j1 c  d. K( h, \
5 A! M8 i+ y! j8 D' E% w- z
Supplementation of EGF-containing growth medium with 20% aqueous humor clearly increased the proliferation of rat RPCs in culture. In addition, it appears that EGF was not responsible for the proproliferative activity of aqueous. First, aqueous supplementation somewhat augmented proliferation in the absence of EGF but not at similar levels. In addition, the activity of aqueous fractions did not suggest a prominent benefit at 6 kDa, which would be consistent with EGF. These same data did not support a role for bFGF either. Thus, the proproliferative activity of EGF, present in the growth medium, and the activity present in aqueous appear to result from different molecules and act synergistically to promote proliferation of rat RPCs.1 k& l8 N9 r: K, E7 v

It will be of considerable interest to know whether EGF- and aqueous-/ascorbic acid-based activities are mediated through different pathways and whether this finding can be extended to RPCs of different species. This is of considerable potential interest in the setting of human RPCs, which exhibit limited ability to passage in culture . Of equal importance is the question of whether exposure of RPCs to aqueous and/or ascorbic acid influences the maturation or fate decisions of the cells when cultured over longer time periods than were possible under current experimental conditions.$ `" r2 [: d- o4 m
- M) v  V9 J5 F; _/ H) N2 B0 o
It will also be of interest to determine whether vitreous liquid exerts a proliferative effect on precursor cells as well and, if so, whether such an effect can be attributed to the same constituents as those present in aqueous humor. Studies of the relationship between progenitor cells and intraocular fluids should lead to a better understanding of ocular development and assist in the development of novel therapeutic modalities based on cell transplantation. Furthermore, the demonstration of potentially novel regulatory factors in the fluid component of the ocular microenvironment may stimulate the search for analogous soluble factors in other stem cell-containing compartments of the body.
+ B; Y0 H* Q- I* v) H9 i& |6 I* |
DISCLOSURES! p0 o- z1 S) d/ A  ^

The authors indicate no potential conflicts of interest.
7 Q, Y, O5 e" `# B) U* d  L
ACKNOWLEDGMENTS

This work was supported by the Danish Eye Foundation, the Novo Nordic Foundation, the Danish Eye Health Society (Værn om Synet), the Synoptic Foundation, and the John and Birthe Meyer Foundation; all support is greatly appreciated. We also thank the Hoag Foundation and Singapore Eye Research Institute for salary support (H.K.). The help and advice of Charly Garbarsh and Carsten Ropke are greatly appreciated.
      【参考文献】. W. i- s; y7 G! N* q

Dowling JE, Sidman RL. Inherited retinal dystrophy in the rat. J Cell Biol 1962;14:73¨C109.
4 H, n/ }+ I0 F4 I$ P
Chaitin MH, Hall MO. Defective ingestion of rod outer segments by cultured dystrophic rat pigment epithelial-cells. Invest Ophthalmol Vis Sci 1983;24:812¨C820./ V; K/ [) r% K$ Q

Mullen RJ, Lavail MM. Inherited retinal dystrophy¡ªPrimary defect in pigment epithelium determined with experimental rat chimeras. Science 1976;192:799¨C801.
) E' b0 S4 p) z; y4 `# {+ C$ L1 r
Villegas-Perez MP, Lawrence JM, Vidal-Sanz M et al. Ganglion cell loss in RCS rat retina: A result of compression of axons by contracting intraretinal vessels linked to the pigment epithelium. J Comp Neurol 1998;392:58¨C77.0 l% p3 e# h7 b7 T! _
/ t. S, y4 j. S
D'Cruz PM, Yasumura D, Weir J et al. Mutation of the receptor tyrosine kinase gene Mertk in the retinal dystrophic RCS rat. Hum Mol Genet 2000;9:645¨C651.0 i2 R& c1 W. _# X3 X9 ~
% d1 V& l3 o6 }
Lavail MM, Unoki K, Yasumura D et al. Multiple growth-factors, cytokines, and neurotrophins rescue photoreceptors from the damaging effects of constant light. Proc Natl Acad Sci USA 1992;89:11249¨C11253.0 M2 X, ~  ], K- X
( I! R6 b, r6 ]6 M! _* o. ~
Anderson RE, Maude MB, McClellan M et al. Low docosahexaenoic acid levels in rod outer segments of rats with P23H and S334ter rhodopsin mutations. Mol Vis 2002;8:351¨C358." r6 A7 K) r% Q, V7 r& _1 m
5 J5 I* w$ O& }! l6 Y
Grozdanic SD, Betts DM, Sakaguchi DS et al. Temporary elevation of the intraocular pressure by cauterization of vortex and episcleral veins in rats causes functional deficits in the retina and optic nerve. Exp Eye Res 2003;77:27¨C33.

Faktorovich EG, Steinberg RH, Yasumura D et al. Photoreceptor degeneration in inherited retinal dystrophy delayed by basic fibroblast growth-factor. Nature 1990;347:83¨C86.6 L* @7 h$ I& j$ d$ I: M! R4 z4 Q5 ^& ~
( X! w# v: t& W
Faktorovich EG, Steinberg RH, Yasumura D et al. Basic fibroblast growth-factor and local injury protect photoreceptors from light damage in the rat. J Neurosci 1992;12:3554¨C3567.

Whiteley SJ, Klassen H, Coffey PJ et al. Photoreceptor rescue after low-dose intravitreal IL-1beta injection in the RCS rat. Exp Eye Res 2001;73:557¨C568.# i% h( u4 ?* J5 t
' S+ c1 h. ^9 Z: o3 X
Li LX, Turner JE. Inherited retinal dystrophy in the RCS rat¡ªPrevention of photoreceptor degeneration by pigment epithelial-cell transplantation. Exp Eye Res 1988;47:911¨C917.: t0 g  x8 i0 X- @0 x7 F
7 c* F- p3 J' u$ u
Sauve Y, Klassen H, Whiteley SJO et al. Visual field loss in RCS rats and the effect of RPE cell transplantation. Exp Neurol 1998;152:243¨C250.4 z& T) H( s, I7 p6 p4 \% P
) t" L8 f6 }5 p( o
Klassen H, Whiteley SJO, Young MJ et al. Graft location affects functional rescue following RPE cell transplantation in the RCS rat. Exp Neurol 2001;169:114¨C121.% ?% F  D$ B, |$ l

Gouras P, Lopez R, Kjeldbye H et al. Transplantation of retinal epithelium prevents photoreceptor degeneration in the RCS rat. Prog Clin Biol Res 1989;314:659¨C671.3 M/ `/ Y2 R; b+ n7 u7 X
) J3 F4 J) }7 {% M
Lawrence JM, Sauve Y, Keegan DJ et al. Schwann cell grafting into the retina of the dystrophic RCS rat limits functional deterioration. Invest Ophthalmol Vis Sci 2000;41:518¨C528.

Chacko DM, Rogers JA, Turner JE et al. Survival and differentiation of cultured retinal progenitors transplanted in the subretinal space of the rat. Biochem Biophys Res Commun 2000;268:842¨C846." I9 D0 O4 |& n9 `" z

Young MJ, Ray J, Whiteley SJO et al. Neuronal differentiation and morphological integration of hippocampal progenitor cells transplanted to the retina of immature mature dystrophic rats. Mol Cell Neurosci 2000;16:197¨C205.
* @) k1 r+ |' X. L5 Y' d2 B% A; y
Nishida A, Takahashi M, Tanihara H et al. Incorporation and differentiation of hippocampus-derived neural stem cells transplanted in injured adult rat retina. Invest Ophthalmol Vis Sci 2000;41:4268¨C4274.5 O% ~) n( B5 P" e' L  p5 f1 ]* A: z
" \8 ^/ ~8 Z5 w+ [; k9 P
Warfvinge K, Kamme C, Englund U et al. Retinal integration of grafts of brain-derived precursor cell lines implanted subretinally into adult, normal rats. Exp Neurol 2001;169:1¨C12.
9 h+ O6 g9 R6 E+ N* _* w
Kurimoto Y, Shibuki H, Kaneko Y et al. Transplantation of adult rat hippocampus-derived neural stem cells into retina injured by transient ischemia. Neurosci Lett 2001;306:57¨C60.

Takahashi M, Palmer TD, Takahashi J et al. Widespread integration and survival of adult-derived neural progenitor cells in the developing optic retina. Mol Cell Neurosci 1998;12:340¨C348.

Van Hoffelen SJ, Young MJ, Shatos MA et al. Incorporation of murine brain progenitor cells into the developing mammalian retina. Invest Ophthalmol Vis Sci 2003;44:426¨C434.
8 }* c6 V; u4 j* N! t5 l8 S
Klassen HJ, Ng TF, Kurimoto Y et al. Multipotent retinal progenitors express developmental markers, differentiate into retinal neurons, and preserve light-mediated behavior. Invest Ophthalmol Vis Sci 2004;45:4167¨C4173.

Ahmad I, Dooley CM, Thoreson WB et al. In vitro analysis of a mammalian retinal progenitor that gives rise to neurons and glia. Brain Res 1999;831:1¨C10.
, d5 G* z" ]. b6 S7 v  \. k, s
Yang P, Seiler MJ, Aramant RB et al. Differential lineage restriction of rat retinal progenitor cells in vitro and in vivo. J Neurosci Res 2002;69:466¨C476.
8 P" V! w! ]# U) t% f( A) N
Marmor MF. Control of subretinal fluid: Experimental and clinical studies. Eye 1990;4:340¨C344.

Lai YL, Fong D, Lam KM et al. Distribution of ascorbate in the retina, subretinal fluid and pigment epithelium. Curr Eye Res 1986;5:933¨C938.
" B  o/ M: M2 J/ h
Litin BS, Jones MA, Herschler J. Aqueous humor-stimulated protein biosynthesis in ocular tissue fibroblast culture. Exp Eye Res 1985;41:183¨C189.

Litin BS, Jones MA, Herschler J. Heat and protease treatment of aqueous humor: Effect on cell DNA synthesis and growth. Graefes Arch Clin Exp Ophthalmol 1985;222:154¨C157.
: j9 x0 ^4 c' B, s9 [
Weinsieder A, Reddan J, Wilson D. Aqueous humor in lens repair and cell proliferation. Exp Eye Res 1976;23:355¨C363.
, u- k; M, h5 r# A+ h
Reddan JR, Weinsieder A, Wilson D. Aqueous humor from traumatized eyes triggers cell division in the epithelia of cultured lenses. Exp Eye Res 1979;28:267¨C276.

Ledbetter SR, Hatchell DL, O'Brien WJ. Differential effect of secondary aqueous humor from rabbit and cat on DNA synthesis of cultured bovine corneal endothelial cells. Curr Eye Res 1981;1:723¨C726.* }( |1 \+ M0 M& |
3 k6 ~4 v. _2 E) P6 X, E
Ledbetter SR, Hatchell DL, O'Brien WJ. Secondary aqueous humor stimulates the proliferation of cultured bovine corneal endothelial cells. Invest Ophthalmol Vis Sci 1983;24:557¨C562.2 S% f& q& b0 a6 V& C6 U. A

Herschler J. The inhibitory factor in aqueous humor. Vision Res 1981;21:163.

Albrink WS, Wallace AC. Aqueous humor as a tissue culture nutrient. Proc Soc Exp Biol Med 1951;77:754¨C758." V" F2 t- @) k

Herschler J, Claflin AJ, Fiorentino G. The effect of aqueous humor on the growth of subconjunctival fibroblasts in tissue culture and its implications for glaucoma surgery. Am J Ophthalmol 1980;89:245¨C249.
' ^7 ~* L2 |8 ]0 r5 _
Radius RL, Herschler J, Claflin A et al. Aqueous humor changes after experimental filtering surgery. Am J Ophthalmol 1980;89:250¨C254.

Burke J, Foster S, Herschler J. Aqueous humor as a modulator of growth in fibroblast cultures. Curr Eye Res 1982;2:835¨C841.: X, d: P9 j% E: O8 b) c- ?7 k0 j

Saari KM, Aine E, Parviainen MT. Determination of protein content in aqueous humour by high-performance gel filtration chromatography. Acta Ophthalmol (Copenh) 1983;61:611¨C617.
: Y: Y/ n. U6 Q6 Q5 a7 o. R% u
Dernouchamps JP, Heremans JF. Molecular sieve effect of the blood-aqueous barrier. Exp Eye Res 1975;21:289¨C297.
  g; m; U5 P6 u* Q2 W6 t$ p1 m
Grabner G, Zehetbauer G, Bettelheim H et al. The blood-aqueous barrier and its permeability for proteins of different molecular weight. Albrecht Von Graefes Arch Klin Exp Ophthalmol 1978;207:137¨C148.

Fielder AR, Rahi AH. Immunoglobulins of normal aqueous humour. Trans Ophthalmol Soc U K 1979;99:120¨C125.' e; g7 z1 L; N. ^9 [

Gaasterland DE, Pederson JE, MacLellan HM et al. Rhesus monkey aqueous humor composition and a primate ocular perfusate. Invest Ophthalmol Vis Sci 1979;18:1139¨C1150.

Taylor A. A review of the influence of aqueous humor on immunity. Ocul Immunol Inflamm 2003;11:231¨C241.$ \3 m/ P& A, j, {# ~
1 k1 h5 C8 ?5 ?4 e9 I
Tsuneto M, Yamazaki H, Yoshino M et al. Ascorbic acid promotes osteoclastogenesis from embryonic stem cells. Biochem Biophys Res Commun 2005;335:1239¨C1246.; `6 i) I' g9 A  c* G3 A

Deliloglu-Gurhan SI, Vatansever HS, Ozdal-Kurt F et al. Characterization of osteoblasts derived from bone marrow stromal cells in a modified cell culture system. Acta Histochem 2006;108:49¨C57." [/ W" d+ @/ c# H
1 ~6 i# [- R% A0 w: o
Carinci F, Pezzetti F, Spina AM et al. Effect of vitamin C on pre-osteoblast gene expression. Arch Oral Biol 2005;50:481¨C496.
+ S# t* {+ B# c4 a6 f+ v- \6 K
Smith J, Ladi E, Mayer-Proschel M et al. Redox state is a central modulator of the balance between self-renewal and differentiation in a dividing glial precursor cell. Dev Biol 2000;97:10032¨C10037.

Lee J, Chang MY, Park CH et al. Ascorbate-induced differentiation of embryonic cortical precursors into neurons and astrocytes. J Neurosci Res 2003;73:156¨C165.& _  \: Q8 z3 u! e7 W; G, s

Klassen H, Ziaeian B, Kirov II et al. Isolation of retinal progenitor cells from post-mortem human tissue and comparison with autologous brain progenitors. J Neurosci Res 2004;77:334¨C343.
6 `$ |2 r' n2 N; _8 x  a+ `
Raff MC, Williams BP, Miller RH. The in vitro differentiation of a bipotential glial progenitor cell. EMBO J 1984;3:1857¨C1864.

收藏0 分享0 顶0 踩0