Descriptive and quantitative study of ultrastructural changes in the apical meristem of mustard in transition to flowering: II. the cytoplasm, mitochondria and proplastids

The present report is a part of a quantitative investigation of the ultrastructural changes accompanying switching of a group of cells to a new morphogenetic sub-programme involving a dedifferentiation process in higher plants. The model system selected for this study was the transition from leaf initiation to flower production at the shoot apical meristem of Sinapis alba (mustard), a long-day species requiring a single long day for induction of flowering.

The cytoplasm, ^* mitochondria and proplastids are covered here. The nucleus and its components were covered in the first paper of this series (Havelange & Bernier, 1974). Other changes will be considered in further papers.

Both the planimetric and the point-counting stereological methods were used for the quantitative evaluation of subcellular changes and it was shown that both methods yielded very similar results and were equally appropriate for the study (Havelange & Bernier, 1974). Only results of a planimetric analysis are given in the present report.

Sixty-five-day-old vegetative plants of Sinapis alba L., grown in 8-h days, were subjected to a single 22-h day, and were then returned to 8-h days. The meristems of these plants will be referred to as ‘evoked’ in contrast to the meristems of control vegetative plants continuously kept in 8-h days. Control meristems were collected at 3 different times during a 24-h cycle. Evoked meristems were collected at 18, 22, 26, 30, 38, 46, 54, 62, and 72 h after the beginning of the long day. Growing conditions for the plants, methods for the preparation of sections, sampling, and statistical analysis of the results have been reported (Havelange & Bernier, 1974). Each value in Figs. 1–5 and Table 1 is an average of 80 cells from 4 different meristems.

The cross-sectional areas of the cell (without cell wall), nucleus, and of all individual mitochondrion, proplastid, and vacuole profiles were measured with a planimeter. The cross-sectional area of the cytoplasm was determined by subtracting the areas of the nucleus, mitochondria and proplastids from the area of the cell. The cross-sectional area of the cytoplasmic matrix was obtained by subtracting the area of the vacuoles from the area of the cytoplasm. The length of the ER was measured with a curvimeter.

In each photographic print, the numbers of mitochondrion, proplastid, vacuole, and dictyosome profiles were determined. These counts cannot be related to the number of particles in a unit volume of tissue without some knowledge or assumptions of particle shape (Weibel, Kistler & Scherle, 1966) and they will be given in this paper only as preliminary indicative results.

Succinic dehydrogenase was localized using the method of Nachlas et al. (1957). Fresh free-hand longitudinal sections of meristems were placed in a solution of sodium succinate and nitro BT in phosphate buffer at pH 7·6. After 40-min incubation at 37 °C in the dark, sections were rapidly washed in 50 % ethanol and mounted in glycerin jelly. Control sections were incubated in solutions with sodium malonate added or with sodium succinate deleted. Heat inactivated tissue sections were also used as controls.

The 2 main cytoplasm components were the cytoplasmic matrix and the vacuolar apparatus. The cross-sectional area of the cytoplasmic matrix was constant throughout a 24-h cycle in cells of control meristems. The size of this cell parameter increased in evoked meristems. A first maximum in size was reached at 26 h after the start of the long day and a second at 54 h (Fig. 1).

The vacuoles of the meristematic cells of mustard were of a variable size and shape, and contained various inclusions and deposits (Fig. 6). The decrease of cellular and membranous inclusions in the vacuoles of evoked meristems (Fig. 7) was described in a preliminary partial report (Havelange, 1972). The cross-sectional area of the vacuolar apparatus considered as a whole was identical in all batches of both control and evoked meristems (Fig. 1). However, as the size of the cytoplasmic matrix increased in evoked meristems, the vacuoles: cytoplasm ratio decreased in these meristems after 26 h (Table 1). Furthermore there were dramatic quantitative changes at the level of individual vacuole profiles. Figs. 2 and 7 show that the number of vacuole profiles in evoked meristem cell sections rose abruptly. The rise was already significant at 18 h, but was more marked after 22 h. A sharp reduction in the cross-sectional area of individual vacuole profiles paralleled this increase in the number of vacuole profiles (Figs. 2 and 7).

The ribosomes were a major component of the cytoplasmic matrix. The quantitative evaluation of ribosome number and density will be reported in a further paper. The changes in 2 minor components of the cytoplasmic matrix, the endoplasmic reticulum (ER) and dictyosomes, will be analysed here.

The meristematic cells had very little ER occurring in short randomly oriented strands. The ER was generally associated with ribosomes (Fig. 6). Curvimetric measurements indicated that the ER length in cell cross-section was similar in control and evoked meristems except at 54 h where it was significantly higher in evoked meristems than in control ones (Fig. 3).

There were very few dictyosomes in the cells of control meristems. The mean number per cell section was 0·1. In evoked meristems, collected at 18, 26, 38, 54, 62, and 72 h, this value rose to 0·2, 0·3, 0·4, 0·7, 2·1, and 0·5 respectively. The dictyosomes had 2–6 straight cisternae displaying few small vesicles (Fig. 6).

Mitochondrial profiles of meristematic cells were spherical or ellipsoidal and contained poorly developed vesiculate cristae. They were commonly from 0·6 to 1 μm in diameter. The proplastids, on the contrary, exhibited a great variety of shape and size (from 0·5 to 3 μm in length) even in a single cell. The matrix of the proplastids was denser than that of the mitochondria but had a few scattered light areas in it. Some proplastid profiles were devoid of starch but others contained this material in variable amount. The smallest proplastid profiles had no internal membranes and were thus quite undifferentiated. Most proplastids, however, showed beginnings of lamellar organization and the largest were forming stacked stroma thylakoid systems, but no grana thylakoid. There did not appear to be obvious changes in mitochondrion and proplastid shape and structure during the morphogenetic switch of the meristem. The mitochondria and proplastids of a single cell will be collectively designated hereafter by the terms ‘chondriome’ and ‘plastidome’, respectively.

Planimetric measurements indicated that there was no daily fluctuation in the chondriome cross-sectional area in cells of control meristems and that this cell parameter increased progressively and tremendously in cells of evoked meristems (Fig. 4). The rise was already significant 18 h from the start of the long day. A maximum size was reached at 62 h. This increase in chondriome size could be explained either by an increase in the mean size of individual mitochondrion profiles or by an increase in the number of mitochondrion profiles. The data of Fig. 5 show that the size of individual mitochondrion profiles did not change very much in evoked meristems but that the number of mitochondrion profiles per cell section increased considerably during the transition to flowering. The number of mitochondrion profiles per cell section was about three times higher in evoked meristems collected at 62 h than in control vegetative meristems.

Succinic dehydrogenase was localized in control and evoked meristems. Enzyme activity was present at all stages of the morphogenetic switch. It was low in control vegetative meristems and high in evoked meristems, particularly in those collected at 54 h. Control sections incubated in a medium lacking succinate exhibited faint staining probably caused by endogenous substrate present in the cellular pool. Other control sections exhibited no staining.

Measurements showed that the plastidome cross-sectional area was constant throughout a daily cycle in the cells of control meristems and that it slightly increased in evoked meristems (Fig. 4). The only value in evoked meristems which was significantly higher than in control meristems was obtained at 54 h after the start of the long day. Consequently the plastidome:chondriome ratio decreased markedly during the flower morphogenetic process (Table 1).

A marked accumulation of starch in proplastids occurred in evoked meristems and was followed by starch utilization and disappearance. These changes in starch content will be studied in detail in a further paper.

The pattern of change in size of the cytoplasmic matrix (Fig. 1) was strikingly parallel to that of the nucleus size (Havelange & Bernier, 1974). However, the nucleus-.cytoplasm ratio decreased from values in the range of 0·71 to 0·65 in cells of control meristems and of evoked meristems collected before 46 h after the start of the long day to a value of 0·47 in evoked meristems collected at 62 h (Table 1). This fall implied that growth of the cytoplasm (which is attributable to the growth of the cytoplasmic matrix) was more important than growth of the nucleus after 46 h.

Basically the cytoplasm components, mitochondria and proplastids of the meristematic cells of mustard are similar to those of meristematic tissues of other species (Buvat, 1958; Lance, 1958; Whaley, Mollenhauer & Leech, 1960; Bowes, 1965; Gifford & Stewart, 1967; Clowes & Juniper, 1968) and, except for vacuoles, there do not appear to be obvious qualitative changes in their structure as the meristem becomes reproductive.

The cytoplasmic matrix reaches a maximal size at two times, 26 and 54 h, just prior to 2 mitotic waves which occurred respectively at 26-30 and 62 h (Bernier, Kinet & Bronchart, 1967). That the cytoplasmic matrix has the greatest size just before mitosis is an expected result, since it is well known that it is of maximal size at prophase (Mitchison, 1971). This increase of the cytoplasmic matrix is probably accompanied by an increase in ribosome number. Indeed, although ribosome counts were not given here, it was previously shown with cytophotometric methods that there is an increase in the content of ribosomal components in cells of evoked meristems of mustard (Jacqmard, Miksche & Bernier, 1972). An increase in ribosome density was reported in evoked meristems of Pharbitis and Perilla (Healey, 1964; Nougarède Bronchart, 1965).

The increases in ER amount and dictyosome number which are found in evoked meristems collected just prior to or at the time of the mitotic wave of the 62nd hour are also probably related to mitotic activation of the tissue, as there is evidence that these 2 cytoplasmic components may be concerned in cell plate and new wall formation during cytokinesis (Porter & Machado, 1960; Frey-Wyssling & Mühlethaler, 1965; Pickett-Heaps & Northcote, 1966; Burgess & Northcote, 1968; Hepler & Jackson, 1968). Similar observations were reported in the meristems of 2 other species during transition to flowering (Healey, 1964; Gifford & Stewart, 1965).

The most noticeable change concerning the vacuolar apparatus is replacement of the large vacuoles of the cells of control meristems by an increased number of smaller vacuoles in the cells of evoked meristems. The validity of this observation may be questioned since it is only substantiated by simple counts of vacuole profiles which cannot be related to the true numbers of vacuoles (see Material and methods). However, these counts are believed to reflect the exact situation because identical observations were made with both the light and electron microscopes (thus in both thick and thin sections) in the evoked meristems of all species so far examined in this respect (Buvat, 1952; Lance, 1957; Gifford & Tepper, 1962; Nougarède, Bronchart, Bernier & Rondet, 1964; Nougarède Bronchart, 1965) and in all tissues involved in a dedifferentiation process associated with a morphogenetic switch (Buvat, 1944-45; Kadej & Rodkiewicz, 1972; Karas & McCully, 1973; Ross, Thorpe & Costerton, 1973). In spite of this great generality the exact function of this splitting of vacuoles (which undoubtedly results in a marked increase of the tonoplast area) in the dedifferentiation process is completely unknown.

This event is already detectable at 18 h after the start of the long day. However, the decrease of the vacuoles-.cytoplasm ratio (i.e. the vacuolation of the cytoplasm), which is exclusively due to the increase in size of the cytoplasmic matrix, seems to be a later event, occurring after 26 h. The vacuolation of plant cells is classically attributed to the fact that the plant cannot produce sufficient protein to fill all its cells with cytoplasmic matrix. Instead, the available space is filled by the watery sap of the vacuole(s) (Frey-Wyssling & Mühlethaler, 1965). The present data are consistent with this view since, according to the prediction, the decrease in vacuolation is paralleled by an increase in size of the cytoplasmic matrix.

The increase in size of the chondriome at the time of floral evocation, as well as the parallel increase in succinic dehydrogenase activity, certainly reflects a rise in cellular respiratory activity. Obviously the morphogenetic switch of the meristem requires large amounts of energy, and since the meristem is essentially a non-photosynthetic tissue, the production of this energy is mainly through respiration. This energy could come from the large amounts of soluble sugars that are produced in the leaves exposed to an inductive long day and subsequently translocated to the apical meristem (M. Bodson, unpublished result).

The increase in chondriome size during transition to flowering has never been reported before, but a rise in respiratory activity was found in evoked meristems of 3 other species (Thein, 1957; Teltscherová Krekule, 1966; Opatrná, 1970; Klopfer, 1973).

As a result of this increase in chondriome size in cells of evoked meristems, there is a marked decrease of the plastidome:chondriome ratio, and thus a profound disturbance in the balance that exists between these 2 organelle families in the cells of control meristems. As the increase in respiratory activity is a common feature of most morphogenetic systems in plants, it would be interesting to know if the decrease of the plastidome:chondriome ratio is also a universal event of these systems.

An interesting fact concerning the increase in chondriome size is that it is already detectable 18 h after the start of the long day, i.e. at the estimated time of arrival of the leaf-generated floral stimulus at the meristem (Kinet, Bodson, AJvinia & Bernier, 1971). The only other structural change that is also detectable at this time is the splitting of vacuoles. Remarkably enough, these 2 changes have in common that they both lead to an increase in membrane area and are thus concerned with increased membrane synthesis. In this respect this very early effect of the floral stimulus is very similar to that produced by various animal and plant hormones in their respective target tissues (Hokin, 1968; Jones, 1969; Mills & Topper, 1970).

Fig. 5 shows that the increase in chondriome size is due to an increase in the number of mitochondrial profiles, suggesting that there is a multiplication of mitochondrial particles in cells of evoked meristems. However, since the exact shape of these particles is not known, this tentative suggestion has to be confirmed by further work.

The decrease in the nucleus: cytoplasm ratio that has been found in the evoked meristems of mustard is a completely new and unexpected result. This change is probably not essential since it occurs after the transforming meristem has reached a point of no return at 44 h after the start of the long day (Kinet et al. 1971).

This work was supported by the F.R.F.C. (Fonds de la Recherche Fondamentale Collective) of Belgium, grant no. 10118.

The authors wish to express their thanks to Professor R. Bronchart, University of Liege, Belgium, for his continuous interest, and Professor W. M. Laetsch, University of California, Berkeley, U.S.A., for reviewing the manuscript.