y. Cell Sd. 83, 213-221 (1986)
213
Printed in Great Britain © The Company of Biologists Limited 1986
THE OCCURRENCE OF LAMPBRUSH CHROMOSOMES
IN EARLY DIPLOTENE OOCYTES OF XENOPUS
LAEVIS
E. L. D. MITCHELL AND R. S. HILL
Department of Genetics, University of Liverpool, P.O. Box 147, Liverpool L69 3BX, UK
SUMMARY
Transmission electron microscopy has been used to study the chromosomes found in the early
diplotene oocytes of Xenopus laevis. Chromosomes from 40 /Urn early diplotene oocytes were found
to possess a normal lampbrush chromosome morphology. The contour length of the loops found on
these chromosomes was measured, the mean value determined and compared with that for
lampbrush chromosomes found on 300/xm mid-diplotene (Dumont stage II) oocytes. The mean
contour length of the loops from 40 fim oocytes (12-33 ± 6-5 jim) was longer than that for the
300 ^m oocytes (7-897 ± 5-22/im). Analysis of variance showed these two values to be significantly
different (P<0-01). The mean loop density (number of loops per 25/im chromosome axis) was
also determined for the two size classes of oocytes, being 9-33 ± 2-05 per 25 fjm of chromosome axis
for the 40 /im size class and 11 -2 ± 2-435 per 25 ^m of chromosome axis for the 300 /im size class.
Analysis of variance showed these values to be significantly different (P< 0-01). The results of the
present study demonstrate that the lampbrush chromosomes found in 40 fim early diplotene
oocytes possess a greater loop contour length and a lower density of loops than those found in
300 fim, mid-diplotene oocytes. Possible reasons for these observations are discussed.
INTRODUCTION
Lampbrush chromosomes are found in the oocytes of most animal species studied
and take their name from the lateral loops, which extend from the chromosome
axis (Macgregor, 1977). The loops are the site of transcriptional activity and the
chromosomes persist in this form throughout the extended diplotene phase of the
first meiotic division.
Dumont (1972), using cytological sections, studied the morphology of the
chromosomes in Xenopus laevis throughout oogenesis and made a classification of
the oocytes. There are six stages in his classification and the lampbrush chromosomes
are stated to be first found in Dumont stage II oocytes.
Using the Miller spreading technique (Miller & Bakken, 1972) to study A", laevis
lampbrush chromosome development it was shown that lampbrush type transcription is established rapidly and early in diplotene (Hill & Macgregor, 1980). This
correlates with biochemical studies that have shown that various classes of RNA,
especially high molecular weight non-ribosomal RNA, are synthesized in early
diplotene (Thomas, 1974; Van Gansen et al. 1976; Scheer et al. 1976). It has also
been found that the poly(A)+ RNA stored in mature X. laevis oocytes starts to
accumulate in Dumont stage I oocytes (Wakahara, 1981; Rosbash & Ford, 1974).
Key words: lampbrush chromosomes, oocytes, diplotene, Xenopus laevis.
214
E.L.D. Mitchell and R. S. Hill
Furthermore, studies using in vitro wheat germ systems have shown that mRNAs
translatable into proteins are readily extractable from Dumont stage I, Xenopus
oocytes (Darnbrough & Ford, 1976; Ruderman & Pardue, 1977). It appears therefore that a large amount of RNA is synthesized in what is thought to be the prelampbrush phase.
The Hill and Macgregor study yielded quantitative information about the size of
transcription units, the spacing of RNA polymerases and the packaging of RNP
transcripts, and from these data it was concluded that the higher-order structure of
chromatin fibres in the pachytene-early diplotene chromosome is modified before
the onset of intensive RNA transcription. However, because of the low ionic
concentrations and high-alkaline pH conditions used, the Miller spreading technique
completely destroys the in vivo architecture of the chromosomes and unravels the
chromatin fibres, so that it yields little information on the structural organization of
the intact chromosome.
The aim of the present study was to isolate pachytene and early diplotene
chromosomes in an intact state, by modifying the isolation conditions used by Hill
and Macgregor. The modifications involved increasing the salt concentration of the
isolation medium to stabilize the higher-order chromatin structure at neutral pH.
Once the chromosomes were isolated the morphology was investigated using both
phase-contrast and electron microscopy. The morphology of the chromosomes from
the early oocytes was compared with that of normal lampbrushes from mid-diplotene
oocytes (Dumont stage II; Dumont, 1972).
MATERIALS AND METHODS
Isolation of chromosomes from previtellogenic oocytes
Oocytes in early meiotic prophase (approx. 15-60/im diameter) were collected from the ovaries
of young Xenopus laevis, which had recently undergone metamorphosis. The ovaries were treated
with collagenase according to the method of Eppig & Dumont (1976). Small pieces of ovary were
rinsed in Ca2+-free OR2 medium (Eppig & Dumont, 1976) and then digested in 0-4% collagenase
solution (Sigma, type I), made up in Ca2+-free OR2, for 2-4 h. The oocytes, free of contaminating
follicle cells, were sorted according to size.
Oocytes, 5-10 of the same size, were placed in a droplet of 0-25 % of the commercial detergent
Joy (Procter & Gamble, Cincinnati, Ohio) in Isolation Medium (IM: 6mM-KH2PO4, 73 mMNa 2 HPO 4> 78mM-NaCl, 38-8mM-KCl; Muller, 1974) on a piece of Parafilm and allowed to lyse.
After lOmin the lysed cells were transferred to a microcentrifugation chamber. The chamber
contained a lower layer of 1 % glutaraldehyde in 0-1 M-sucrose in IM and an upper layer of IM. The
lysed cells were allowed to disperse for 15min before being centrifuged at 4000 rev. min" 1 for
20 min onto a glow-discharged, Formvar/carbon-coated copper grid. The grids were then rinsed in
0-5% Photoflo (Kodak) for 10s before being air dried and rotary shadowed with platinum/
palladium at an angle of 10°.
Isolation of chromosomes from mid-diplotene oocytes
The germinal vesicles of 300 fim diameter oocytes were manually isolated in a 5:1 saline (5 parts
0-lM-KCl to 1 part 0-lM-NaCl). The germinal vesicle was then transferred to a watchglass
containing clean 5:1 saline and was manually opened using a pair of tungsten needles. The nuclear
sap was then transferred to a microcentrifugation chamber containing an upper layer of 0-25 % Joy
in IM and a lower layer of 1 % glutaraldehyde in 0 - l M-sucrose in IM. The sap was allowed to
Lampbrush
chromosomes in Xenopus
215
disperse for IS min before being centrifuged at 4000 rev. min" 1 for 20 min, onto a glow-discharged
Formvar/carbon-coated copper grid. The grids were then rinsed in distilled water and dehydrated
through SO %, 70 %, 95 % and 100 % (twice) (v/v) ethanol, being left in each ethanol solution for
10 min. They were then critical-point dried for 1 -5 h using CO2, before being rotary shadowed with
platinum/palladium at an angle of 10°. The preparations were examined using a Zeiss EM 10
transmission electron microscope (TEM).
Measurements of the contour lengths of all measurable loops were made using an Apple
Graphics Tablet. Five preparations, prepared under identical conditions, were used to obtain data
for the 300 ^m mid-diplotene oocytes and two preparations were used for 40,um early diplotene
oocytes. These same preparations were also used to determine the loop density (number of loops
per 25 fim chromosome axis) for each oocyte size.
To determine the effect of critical-point drying on loop length, the lengths of the loops of
preparations from the 300 fim oocytes were measured both before and after drying, using phasecontrast microscopy.
Where appropriate the data were tested by analysis of variance.
RESULTS
The method used to isolate chromosomes from early diplotene oocytes provided us
with only two analysable preparations, despite the method being repeated on a
number of occasions; the reasons for this problem will be discussed later. The
conclusions that we can draw from these results therefore must be viewed in the light
of this limited sample.
Fig. 1A is an electron micrograph of a preparation obtained from a 40 fim
diameter, early diplotene oocyte. When compared with Fig. 2A, an electron micrograph of a 300fim diameter oocyte (Dumont stage II), it can be seen that the
chromosomes from the 40 ^m oocyte have a normal lampbrush chromosome morphology, with loops radiating from chromomeres on the chromosome axis. Also, the
loops found on the lampbrush chromosomes of both size classes of oocyte, are
covered with ribonucleoprotein matrix, as seen in Figs IB and 2B, thus demonstrating that the lampbrush chromosomes in the 40 fim early diplotene oocytes are
transcriptionally active.
Fig. 3 shows the distribution of contour loop length for both size classes of oocyte.
The two size classes of oocyte both appear to have the same modal size category of
loop contour length. However, the lampbrush chromosomes of 40 ^m oocytes
possess a greater proportion of larger loops than the chromosomes found in 300 fim
oocytes.
The mean loop contour length of lampbrush chromosomes from five preparations
of the 300 fim size class of oocytes and two preparations from 40 fim size oocytes was
determined. It was found that the mean loop length for the chromosomes in the
40,um oocytes (12-33 ± 6-51 fim) was longer than that for the chromosomes in the
300 jxm oocytes (7-897 ± 5-22 fim) (Table 1). An analysis of variance was carried out
on these results and indicated that there is a significant difference between the two
means (Table 1) (P<0-01).
One possible reason for the difference between the two means could be the
difference in the drying techniques used for the two different-sized oocytes. When
lampbrush chromosomes from 300 fim diameter (Dumont stage II) oocytes were
initially isolated the preparations were dried with the photographic detergent,
216
E. L. D. Mitchell and R. S. Hill
Photoflo. However, it was found that for those preparations made from the larger,
mid-diplotene, 300 jUm diameter oocytes, detergent drying tended to mat the loops
together and therefore rendered measurement of the loop contour length impossible.
Fig. 1. A TEM preparation from a 40 jum diameter, early diplotene oocyte. A. Part of a
chromosome. Bar, 5 fim. B. A chromosome loop. Bar, 2/lm. Note the normal lampbrush
chromosome morphology and the RNP matrix on the loops.
Fig. 2. A TEM preparation from a 300 jUm diameter, mid-diplotene oocyte. A. Part of a
lampbrush chromosome. Bar, 5 /im. B. A typical loop found on such chromosomes. Bar,
2/lm.
Lampbrush chromosomes in Xenopus
10
15
20
25
Loop contour length
30
217
35
Fig. 3. The distribution of loop contour length for both the 40 fim diameter oocytes and
the 300 fan diameter oocytes. (x
x) 40 fim oocytes; (•
• ) 300 fim oocytes.
Critical-point drying the preparations gave much better separation of the loops and
so this method was used when preparing lampbrush chromosomes from 300 fim
diameter oocytes. To determine whether critical-point drying the preparations
caused excessive shrinkage of the loops, phase-contrast micrographs of the preparations, on electron-microscope grids, were taken both before and after drying.
The contour lengths of the loops were measured for both situations and compared
(Table 2). The mean loop lengths were quite similar, 9-78 ± 4-439 fim before drying
Table 1. The mean loop contour length and number of loops per 25 fim of chromosome
axis of the lampbrush chromosomes of the two different size classes ofoocytefrom
X. laevis
Oocyte
class size
(fim)
No. of
preparations
used
No. of
loops
measured
40
341
300
562
Mean loop
contour length
12-33 ±6-51
(P<001)
7-897 ±5-22
No. of loops
per 25 fim of
chromosome axis
9-33 ±2-05
(P<0-01)
ll-2± 2-435
Table 2. The mean loop contour length of the lampbrush chromosomes from a single
300 fim diameter oocyte before and after critical-point drying
Before drying
Mean loop contour length (fim)
Probability - Analysis of variance
After drying
9-78 ±4-439
8-78 ±4-184
P>0-05, not significant
218
E.L.D. Mitchell and R. S. Hill
and 8-78 ± 4-184jUm after drying; analysis of variance showed that the means were
not significantly different (P>0-05). This suggests therefore, that the preparations
of the two different size classes of oocytes are comparable.
The loop density (number of loops per 25 fim of chromosome axis) was determined for both size classes of oocyte and compared (Table 1). The mean values
obtained were different, 9-33 ± 2-05 loops per 25 [im chromosome axis for the 40 ^tm
size oocytes, and 11 -2 ± 2-435 loops per 25 fim chromosome axis for the 300 jitm size
oocytes. An analysis of variance of these means showed that the difference observed
is significant (P<0-01) and this suggests that lampbrush chromosomes from the
40 (im, early diplotene oocytes may have a lower density of loops than those of the
300 ;Um, mid-diplotene oocytes.
DISCUSSION
The results obtained clearly demonstrate that in early diplotene oocytes of 40 [Am
diameter lampbrush chromosome morphology is already established and that the
chromosome loops are formed very early in diplotene. This finding correlates with
results from an earlier Miller spreading study (Hill & Macgregor, 1980) where
chromosomal transcription units, comparable in size to those found in fully formed
lampbrush chromosomes were present in early diplotene nuclei of 25—40/im
diameter. The authors suggested that these transcription units may represent the
earliest loops to emerge and the present results support this suggestion.
The presence of large numbers of transcription units in early diplotene oocytes
suggests that a high level of RNA synthesis is occurring, a fact that is supported by
the findings of a study on tritiated-uridine incorporation into Xenopus oocytes
(Macgregor, 1980). In this study it was shown that the level of RNA synthesis, in
oocytes, increases quite sharply between pachytene and early diplotene and once this
elevated synthetic rate is reached it continues while the oocyte increases in size. The
occurrence of a high level of RNA synthesis in early diplotene oocytes is further
supported by the results of several biochemical investigations, which demonstrated
the presence of poly(A)+ RNA, translatable mRNA and 4S and 5S RNA in
previtellogenic, Dumont stage I oocytes (Rosbash & Ford, 1974; Thomas, 1974;
Darnbrough & Ford, 1976; Ruderman & Pardue, 1977).
According to Dumont (1972), however, lampbrush chromosomes are not present
in Dumont stage I oocytes (50-300jum in diameter), but form in Dumont stage II
cells (300-450[im in diameter). The findings of the present study are therefore not in
agreement with Dumont's scheme, and demonstrate that lampbrush morphology is
formed and becomes active earlier in oogenesis than was previously supposed.
The loop density (number of loops per 25 jum of chromosome axis) has been shown
to differ significantly between lampbrush chromosomes from the two size classes of
oocyte studied, suggesting that more loops are found on the lampbrush chromosomes
of the 300 ^im mid-diplotene oocytes than on the chromosomes found in the 40 fim
early diplotene cells. This suggests, therefore, that an increasing number of loops
may be emerging as oogenesis progresses.
Lampbrush chromosomes in Xenopus
219
On the other hand, however, owing to slight differences in preparation of the two
size classes, as described in Materials and Methods, the axes of the chromosomes
found on preparations from the 40 jUm size class may be more extended than those on
preparations from the 300 fim size class of oocyte. Lysing the early diplotene oocytes
in the commercial detergent, Joy (Procter & Gamble), and then transferring the
droplet to a chamber, after lOmin dispersion, may have resulted in the chromosome
axes being stretched. The viscous nature of the nuclear sap found in oocytes, in the
later stages of oogenesis, will have provided the chromosomes in the 300 jxm oocytes
with some protection from the stresses of transfer into the microcentrifugation
chamber. If the chromosome axes of the lampbrush chromosomes in the early
diplotene oocytes are stretched, then the number of loops found along a particular
length of chromosome axis will be reduced, as compared to the situation in vivo.
The lampbrush chromosomes found in the early diplotene oocytes appear to have
a greater loop contour length than those of the mid-diplotene (300 jUm diameter)
oocytes, although why there should be this difference is unclear.
One possible reason for the length difference could be the different drying
techniques used for the two different sizes of oocytes. However, a comparison of loop
contour length before and after the 300 fim oocyte preparations were critical-point
dried, demonstrated that there was no excessive shrinkage of the loops during
drying, indicating that the two sets of results from the different-sized oocytes were
comparable. However, the previously mentioned viscous nature of the nuclear sap
of the 300 jUm oocytes will protect the lampbrush loops from stretching during
preparation and this may contribute partly to the observed difference in length.
The preparations from the 40 ^m diameter oocytes were detergent dried and as
previously stated this method tends to mat the loops together in preparations from
oocytes of 300/zm diameter. If this were also the case for the 40 //m oocyte preparations then the measurement of loop contour length may be biased towards longer
loops. Also, the determination of loop density will be underestimated since shorter
loops would not be resolvable. However, matting of the loops was not observed in
preparations of lampbrush chromosomes from the early diplotene oocytes. From
Fig. 1A it can be seen that the loops found on these preparations are well dispersed
and easily distinguishable.
Furthermore, we do not think that such matting has seriously biased our data from
the chromosomes of early oocytes. Fig. 3 shows the distribution of contour loop
length for both size classes of oocyte and it can be seen that, for both classes, loops in
all size categories are found. When comparing the two distributions of loop contour
length, it can be seen that the lampbrush chromosomes of 40 ^m oocytes possess a
greater proportion of larger loops than the chromosomes found in 300 fim oocytes.
These two observations suggest, therefore, that the mean loop contour length of the
loops found on the lampbrush chromosomes in 40 ^m oocytes is indeed greater than
for the 300 [im oocytes.
As oogenesis progresses there may be a reduction in the rate of transcription of
some of the lampbrush transcription units. This would result in a decrease in loading
of RNA polymerase and an increase in spacing between the polymerases, which
220
E. L. D. Mitchell and R. S. Hill
would lead to the intervening chromosome axis being packaged into nucleosomes
and possibly higher-order structures. Such packaging will result in a decrease in
transcription-unit length and consequently a decrease in loop length.
One lampbrush loop is not necessarily just one transcription unit, some loops may
include several transcription units, tandemly arrayed and of like or unlike polarities
(Callan & Lloyd, 1960). Throughout oogenesis some transcription units may be
deactivated, but not necessarily whole loops, and the chromatin folded into a higherorder structure. This packaging of the inactive chromatin would result in a decrease
in loop contour length and may contribute to the observed difference in length
between the early and mid-diplotene oocytes.
A study by Macgregor & Andrews (1977) looking at middle repetitive DNA in
Triturus cristatus carnifex, using in situ hybridization techniques, showed that in
small oocytes some loops are found that are not present in larger oocytes. The
authors suggested that the different patterns of labelled loops seen in large and
small oocytes reflects a progressive and continuous lateral extension of part of the
chromomeric material such that tandemly linked families of sequences will be
exposed for transcription at certain stages of oogenesis only.
It could be that throughout oogenesis certain loops emerge as the transcription
units found on them become activated and this emergence may be compensated for
by the regression of other loops. Loops emerging later on in oogenesis may differ in
contour length from those that have regressed and again may contribute to the
observed difference in length between the two size classes of oocyte.
Finally, for various reasons, only two analysable preparations were obtained from
the 40 [im oocytes. First, there will be loss of material during transfer of the droplet
of 0-25 % Joy in IM, into the microcentrifugation chamber. Lysis of the oocytes will
cause the chromosomes to be released into the droplet and these could stick to the
side or end of the pipette and therefore be lost. Such a situation is also found when
Miller spreading is used according to the original method (Miller & Bakken, 1972).
Here oocyte nuclei (germinal vesicles) are lysed in a droplet of water at pH 9 and it
has been noted that a great loss of chromatin occurs when the droplet is transferred to
a microcentrifugation chamber (Hill, unpublished observations).
When making preparations for the electron microscope from larger oocytes,
300 fim or greater in diameter, we found that the chromosomes did not stick very well
to the electron-microscope grids and on only about 20% of the preparations made
were lampbrush chromosomes found. This factor could also contribute to the small
number of preparations obtained for the 40 [im oocytes. The poor adhesion of the
X. laevis chromosomes, to the grids, may be due to the shortness of their loops as
compared to other amphibian species such as T. cristatus carnifex (Macgregor,
1980). Xenopus lampbrush loops have an L factor (measured as the distance from the
loop inflexion to the chromosome axis) of approximately 5^m, whereas loops of
lampbrush chromosomes of Triturus have an L factor of 11 (im.
Furthermore, it was also noted that when making preparations of 40 [im oocytes
the droplet of IM and 0-25 % Joy, in which the oocytes were lysed, spreads over the
Parafilm. This spreading occurs because the Joy reduces the surface tension of the
Lampbrush chromosomes in Xenopus
221
liquid. Spreading of the droplet will effect the efficiency with which one is able to
pick up the contents of the droplet. Consequently, loss of chromosomal material
will occur when transferring the droplet to the microcentrifugation chamber.
Furthermore, chromosomes may stick to the Parafilm and not be transferred into the
chamber.
This research was supported by the Medical Research Council, grant no G82.076531.BO. The
authors thank Mr J. Smith for his assistance with the electron microscopy, Mr C. J. Veltkamp for
critical-point drying the specimens and Mr P. G. Mclntyre for assisting with the printing of the
figures.
REFERENCES
CALLAN, H. G. & LLOYD, L. (1960). Lampbrush chromosomes of crested newts Triturus cristatus
(Laurenti). Phil. Trans. R. Soc. Lond. B 243, 135-319.
DARNBROUGH, C. & FORD, P. J. (1976). Cell-free translation of messenger RNA from oocytes of
Xenopus laevis. Devi Biol. 50, 285-301.
DuMONT, J. (1972). Oogenesis in Xenopus laevis (Daudin): I. Stages of oocyte development in
laboratory maintained animals. J. Morph. 136, 153-179.
EPPIG, J. J. & DUMONT, J. (1972). A defined nutrient medium for the in vitro maintenance of
Xenopus laevis oocytes. In Vitro 12, 418-427.
HILL, R. S. & MACGREGOR, H. C. (1980). The development of lampbrush chromosome type
transcription in the early diplotene oocytes oiXenopus laevis: An electron microscope analysis.
jf. Cell Set. 44, 87-101.
MACGREGOR, H. C. (1977). Lampbrush chromosomes. In Chromatin and Chromosome Structure
(ed. R. A. Eckhardt & H. Hsueh-jei), pp. 339-357. New York, London: Academic Press.
MACGREGOR, H. C. (1980). Recent developments in the study of lampbrush chromosomes.
Heredity 44(1), 3-35.
MACGREGOR, H. C. & ANDREWS, C. (1977). The arrangement and transcription of "middle
repetitive" DNA sequences on lampbrush chromosomes of Triturus. Chromosoma 63, 109-126.
MILLER, O. L. & BAKKEN, A. H. (1972). Morphological studies of transcription. Ada endocr.
Suppl. 168, 158-177.
MULLER, W. P. (1974). The lampbrush chromosomes oiXenopus laevis (Daudin). Chromosoma
47, 283-296.
ROSBASH, M. & FORD, P. J. (1974). Polyadenylic acid containing RNA inXenopus laevis oocytes.
y. molec. Biol. 85, 87-101.
RUDERMAN, J. V. & PARDUE, M. L. (1977). Cell-free translation analysis of messenger RNA in
echinoderm and amphibian early development. Devi Biol. 60, 48-68.
SCHEER, U., TRENDELENBURG, M. F. & FRANKE, W. W. (1976). Regulation of transcription of
genes of ribosomal RNA during amphibian oogenesis: A biochemical and morphological study.
y. Cell Biol. 69, 465-489.
THOMAS, C. (1974). RNA metabolism in previtellogenic oocytes oiXenopus laevis. Devi Biol. 39,
191-197.
VAN GANSEN, P., THOMAS, C. & SCHRAM, A. (1976). Nucleolar activity and RNA metabolism
in previtellogenic and vitellogenic oocytes of Xenopus laevis: A biochemical and autoradiographical, light and electron microscope study. Expl Cell Res. 98, 111-119.
WAKAHARA, M. (1981). Accumulation, spatial distribution and partial characterization of poly(A) +
RNA in the developing oocytes oiXenopus laevis. y. Embryol. exp. Morph. 66, 127-140.
{Received 4 November 1985 -Accepted 10 February 1986)