Metamorphosis in Xenopus laevis is not associated with large-scale nuclear DNA content variation
Department of Crop Sciences, 320 ERML, 1201 W. Gregory Drive, University of Illinois, Urbana, IL 61801, USA
* Author for correspondence (e-mail: arayburn{at}uiuc.edu)
Accepted 22 September 2004
![]() |
Summary |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: nuclei, flow cytometry, development, metamorphosis, Xenopus
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Fritz et al. (1990) used the
fluorochrome 4-6-diamidino-2-phenylindole (DAPI) in their experiments. DAPI is
a dye that stains DNA and, more specifically, AT-rich DNA
(Shapiro, 1995
). Fritz et al.
(1990
) mentioned that reports
have indicated that DAPI has minor affinity to heterochromatin. Their
experience suggested that the increase in fluorescence intensity observed
using this dye was due to increasing DNA content. Other researchers have
suggested that DAPI is less affected by chromatin condensation than other DNA
dyes (Shapiro, 1995
). Rayburn
et al. (1989
) observed that in
maize nuclear DNA amounts obtained by flow cytometry using the dye DAPI gave
similar results to nuclei stained with Schiff's reagent and analyzed by
microdensitometry. Important to these findings is that the maize lines used in
the study differed substantially in the amount of heterochromatin present.
Thus, the data suggested that DAPI was not affected by heterochromatin.
However, other studies have indicated that DAPI binding to DNA is affected by
chromatin structure (Darzynkiewicz,
1990
). Given the conflicting reports regarding the mechanism of
DAPI, any conclusions on DNA content versus chromatin condensation
using this dye is suspect.
One dye that is known to be affected by chromatin differences is propidium
iodide (PI). PI is a base-pair-intercalating dye. Rayburn et al.
(1992) observed that PI
fluorescence did not reflect reported DNA content differences among maize
lines differing with respect to the amount of heterochromatin each contained.
The authors hypothesized that compacted heterochromatin was not as accessible
to the PI dye. Other investigators have observed that during apoptosis,
chromatin condensation occurs, which results in reduced PI fluorescence
(O'Brien et al., 1998
). It has
now become an acceptable practice to take advantage of the reduced PI
fluorescence due to chromatin compaction to quantify apoptosis by observing
the presence of hypodiploid nuclei using flow cytometry
(Hess et al., 2001
;
Schlosser et al., 2003
;
Souza-Fagundes et al.,
2003
).
The purpose of this study was to use PI, which is well documented to be affected by chromatin changes, to stain nuclei of developing X. laevis tadpoles. Nuclei from the same staged animals were stained in a `native' state directly after nuclear isolation and in an altered state (fixed). Any changes observed in the fluorescence intensity of the stained nuclei between the two treatments would indicate that large-scale DNA variation is not associated with metamorphosis.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The DNA content analysis protocol is a modification of the Gold et al.
(1991) fish cell protocol for
flow cytometry. Tadpoles were homogenized in 10 ml of a 1 x
phosphate-buffered saline (PBS; Sigma, St Louis, MO, USA) and then filtered
through a 53 µm mesh. Four 1 ml aliquots of nuclei suspension per tadpole
were placed in microfuge tubes and microcentrifuged for 8 s at 12,000
g. The supernatants were aspirated and nuclear pellets were
combined for each tadpole by resuspending all four pellets in a total volume
of 800 µl PBS. Microcentrifugation at 12,000 g was
repeated, supernatant aspirated and the resulting pellet resuspended in 800
µl of PBS. The nuclei were then centrifuged in a microfuge for 8 s at
12,000 g, the supernatant aspirated and the nuclear pellet
resuspended in a propidium iodide (PI) stain solution [0.8 g sodium chloride
(137 mmol l1), 0.056 g PIPES, 0.028 g disodium EDTA (0.75
mmol l1), 4 mg deoxyribonuclease-free ribonuclease and 5 mg
propidium iodide in 100 mldistilled water brought to pH 7.50 with NaOH]. The
RNase treatment and staining are both at 4°C due to the fact that at
higher temperatures degradation of the nuclei occurs interfering with the
analysis. After incubating 1 h at 4°C in the dark, the samples were
filtered with a 53 µm mesh. Cell-cycle analysis was done on an EPICS XL
flow cytometer (Coulter Electronics, Hileah, FL, USA) using an excitation
wavelength of 488 nm. 50,000 nuclei were analyzed per sample. The flow rate
resulting in the best most-consistent results was determined by dividing the
nuclei of several animals into three subsamples and analyzing each sample at a
different flow rate (Table 1).
The full peak coefficients of variation (CV) of the G1 peaks were measured and
compared. An ANOVA (analysis of variance) and LSD (least significant
difference) were run using SAS (version 8.2, SAS Institute Inc, Cary, NC,
USA).
|
Stage nuclei samples for fixation were isolated as described above. Nuclei samples were then fixed in 50% ice cold ethanol for 20 min at 20°C. Nuclei suspensions were microcentrifuged for 8 s at 12,000 g and the supernatants removed. The nuclear pellet was then washed in PBS and resuspended in a PI stain solution volume to result in a final concentration of 6 x105 nuclei ml1 of stain. The fixed samples were then placed on ice in the dark for 1 h and filtered through a 53 µm mesh. 10,000 nuclei were analyzed per fixed sample at a flow rate of 200 nuclei per second. The CVs of the G1 peaks were measured and compared. An ANOVA and LSD were run using SAS.
For fixation comparisons, nuclei samples from six tadpoles were divided into subsamples and either fixed in 25%, 50% or 75% ethanol for 20 min at 20°C or left unfixed. Nuclei from each animal were counted, stained at a concentration of 6x105 nuclei ml1 and analyzed as previously described in the fixation method. 10,000 nuclei were analyzed per sample at 200 nuclei per second. The CVs of the G1 peaks were measured and compared. An ANOVA and LSD were run using SAS.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
For our analysis, stages 4851 were clustered in a group denoted as the early two-leg stage (Fig. 1A). The histograms for these animals were observed to have a single peak and to be very consistent from animal to animal (Fig. 2A). At stages 5255 (two-leg stage; Fig. 1B) the single peak began to broaden. In many cases it no longer appeared as a single peak. In a few tadpoles, a second, distinct hyperdiploid peak appears at a higher DNA amount than the previous peak (Fig. 2B). As the tadpole continues through stages 5657 (late two-leg stage), more individuals are observed to have the distinct hyperdiploid peak. At stages 5862 (four-leg tadpole; Fig. 1C) the number of individuals with the hyperdiploid peak decreases and in many of the tadpoles the resolution of the two peaks becomes less evident (Fig. 2C, Table 2).
|
|
As metamorphosis continues, the peaks begin to narrow with the hyperdiploid peak disappearing. At stage 63 (Fig. 1D), the peak symmetry improves (Fig. 2D) and the CV decreases (Table 2). When stage 65 is reached, the peak appears at its best with a very narrow and symmetrical peak and the lowest recorded CV (Table 2). No sign of the hyperdiploid peak is apparent. This nice distinct peak remains until stage 66. As the tadpole completes metamorphosis and develops into a small juvenile frog, the technique used to isolate intact nuclei is no longer effective due to mechanical limitations. Three very distinct LSD groupings were observed among the various stages with respect to CV (Table 2). It should be noted that while presented as discrete units, the stage groupings during metamorphosis are continuous data. Therefore, various types of histograms were observed in each group cluster. The peaks were not necessarily homogeneous within each developmental cluster requiring a large sample size to determine the representative histogram.
Upon using 50% ethanol fixation the CVs of all stages of the developing larvae were improved (Table 3). Only single-peak histograms were obtained (Fig. 3AD). Owing to the presence of only a single peak, fewer animals were required to establish a profile for each stage cluster. Only two overlapping LSD clusters were observed among the stages (Table 3). As in the unfixed nuclei, the best CVs were observed at the four-leg stages. However, no significant difference was observed from the no-legs stage through the early four-leg stage (Table 3).
|
|
Four different fixation methods were compared (Table 4). Nuclei from individual animals were divided into four subgroups. Nuclei from all six animals were analyzed in an unfixed state and fixed at 25%, 50% and 75% ethanol. The largest CV was obtained with the 75% ethanol fixation. The 25% ethanol fixation gave similar CVs to the unfixed nuclei. The best CVs were obtained with 50% ethanol.
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Vassetzky et al. (2000)
reported that chromatin domains do rearrange during X. laevis
development. The domain rearrangement did not appear random but seemed to be
related to the timing of gene expression during development. If such chromatin
rearrangement does occur during larvae development the amount of facultative
heterochromatin may fluctuate. Such fluctuations could result in different
intercalation of the fluorochrome PI within the nucleus of differentiating
tissues. This would result in the DNA amounts appearing different at various
stages due to the heterogeneity of nuclei of different organs or tissue. This
heterogeneity of nuclei could be reflective of differential timing of
metamorphosis in specific tissue/organs. Also, the nuclear heterogeneity could
well be due to the presence of alternative mechanisms of metamorphosis not
involving chromatin changes (Sachs and
Shi, 2000
). Regardless of the mechanism involved in nuclear
heterogeneity within metamorphosing tadpoles, the data presented indicates
that large-scale DNA variation is not involved in metamorphosis.
The results of this study are consistent with DNA content variation within individuals not being associated with metamorphosis. Holtfreter and Cohen (1990) made an interesting observation in hemopoietic frog cells. Studies in their laboratory indicated that the chromatin of erythrocytes had differential accessibility to PI than the chromatin of the leukocytes. Upon fixation with 50% ethanol, the erythrocytes and leukocytes were observed to have the same PI fluorescence. The fixation step resulted in the homogeneity of staining between the erythrocytes and leukocytes. Thus, if the DNA differences observed in the X. laevis developing larvae disappeared, one would expect that large-scale nuclear DNA content variation would not be involved in metamorphosis. Upon fixation the differential PI fluorescence is lost. All of the nuclei appear homogeneous with respect to PI fluorescence intensity.
The ethanol fixation used here has never been reported to result in
selective degradation of DNA. On the contrary, 50% ethanol fixation has been
suggested as the fixation of choice when DNA analysis requires a balance of
homogeneity of staining and high resolution of staining
(Poulin et al., 1994). Ethanol
mechanism of action has not been associated with DNA changes. Reports have
indicated that varying the concentration of ethanol in the fixative from 50%
results in increased variation in the fluorescence of nuclei stained for DNA
content due to fluctuations in chromatin density
(Shapiro, 1995
). Thus in the
present study, if homogeneity of the nuclei distribution is affected by
altering the concentration of ethanol in the fixative, evidence for
fluctuations in nuclear DNA content being involved in metamorphosis would be
weakened. The results presented here clearly demonstrate that deviations from
50% ethanol fixation results in increasing CVs which is indicative of
heterogeneity of the nuclei due to changes other than DNA content changes.
With the differential fluorescence of the larvae being associated with development not observed after optimum fixation (which results in homogenous nuclei and reduces artifacts), it appears clear that the differential fluorescence due to the presence of hyperdiploid nuclei is not the result of large-scale total nuclear DNA variability. Small amplification events too small to be detected by flow cytometry could still be occurring. However, such events could not be responsible for the large amount of fluorescence variability observed in the nuclei of metamorphosing individuals.
![]() |
Acknowledgments |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Birstein, V. J., Poletaev, A. I. and Goncharov, B. F. (1993). DNA content in Eurasion sturgeon species determined by flow cytometry. Cytometry 14,377 -383.[Medline]
Bonaly, J., Bre, M. H., Lefort, T. M. and Mestre, J. C. (1987). A flow cytometric study of DNA staining in situ in exponentially growing and stationary Euglena gracilis.Cytometry 8,42 -45.[Medline]
Darzynkiewicz, Z. (1990). Acridine orange as a molecular probe in studies of nucleic acids in situ. In Flow cytometry and sorting (ed. M. R. Melamed, P. F. Mulvaney and M. L. Mendelsohn), pp. 285-316. New York: John Wiley & Sons.
Fritz, B., Fritz, J. P. and Ulrich, W. (1990). Flow cytometric determination of changes in cellular DNA content during development of Xenopus laevis. In Cytogenetics of Amphibians and Reptiles (ed. E. Olmo), pp.129 -133. Canterbury, UK: World Congress of Herpetology.
Gold, J. R., Ragland, C. J., Birkner, M. C. and Garrett, G. P. (1991). Simple procedure for long-term storage and preparation of fish cells for DNA content analysis using flow cytometry. Progressive Fish Culturist 53,108 -110.[CrossRef]
Hess, K. L., Johnson, J. D. and Cook-Mills, J. M.
(2001). Different orders for acquisition of apoptotic
characteristics by leukocytes. J. Leuk. Biol.
70,405
-412.
Holtfeter, H. B. and Cohen, N. (1990). Fixation-associated quantitative variations of DNA fluorescence observed in flow cytometric analysis of hemopoietic cells from adult diploid frogs. Cytometry 11,676 -685.[Medline]
Nieuwkoop, P. D. and Faber, J. (1956). Normal table of Xenopus laevis (Daudin). Amsterdam: North Holland Publishing Company
O'Brien, I. E. W., Murray, B. G., Baguley, B. C., Morris, B. A. M. and Ferguson, I. B. (1998). Major changes in chromatin condensation suggest the presence of an apoptotic pathway in plant cells. Exp. Cell Res. 241,46 -54.[CrossRef][Medline]
Poulin, N. M., Matthews, J. B., Skov, K. A. and Palcic, B.
(1994). Effects of fixation method in image cytometric
measurement of DNA content and distribution in cells stained for fluorescence
with propidium iodide. J. Histochem. Cytochem.
42,1149
-1156.
Rayburn, A. L., Auger, J. A., Benzinger, E. A. and Hepburn, A. G. (1989). Detection of intraspecific DNA variation in Zea mays L. by flow cytometry. J. Exp. Bot. 40,1179 -1183.
Rayburn, A. L., Auger, J. A. and McMurphy, L. M. (1992). Estimating percentage constitutive heterochromatin by flow cytometry. Exp. Cell Res. 198,175 -178.[Medline]
Sachs, L. M. and Shi, Y. B. (2000). Targeted
chromatin binding and histone acetylation in vivo by thyroid hormone
receptor during amphibian development. Proc. Natl. Acad. Sci.
USA 97,13138
-13143.
Schlosser, S. F., Schuler, M., Berg, C. P., Lauber, K.,
Schulze-Osthoff, K., Schamhl, F. W. and Wesselborg, S.
(2003). Ribavirin and alpha interferon enhance death
receptor-mediated apoptosis and caspase activation in human hepatoma cells.
Antimicrob. Agents Chemother.
47,1912
-1921.
Shapiro, H. M. (1995). Practical Flow Cytometry. New York: Wiley-Liss, Inc.
Souza-Fagundes, E. M., Gazzinelli, G., Parreira, G. G., Martins-Filho, O. A., Amarante-Mendes, G. P., Correa-Oliveria, R. and Zani, C. L. (2003). In vitro activity of labdene diterpene from Alomia myriadenia (Asteraceae): immunosuppression via induction of apoptosis in monocytes. Int. Immunopharmocol. 3, 383-392.[CrossRef]
Tiersch, T. R. and Wachtel, S. S. (1991). On the evolution of genome size of birds. J. Hered. 82,363 -368.[Medline]
Vassetzky, Y., Hair, A. and Mechali, M. (2000).
Rearrangement of chromatin domains during development in Xenopus.Genes Dev. 14,1541
-1552.