Diversity of cell lengths in terminal portions of roots:
implications to cell proliferation
Lance S. Evans
Laboratory of Plant Morphogenesis,Biological Sciences Research Laboratories,Manhattan College,The Bronx,New York,NY,
10471,USA
Received 8 September 1999; received in revised form 8 November 1999; accepted 12 November 1999
Abstract
Terminal meristems are responsible for all primary growth of roots. It has been asserted that all cells of root meristems are actively dividing (no cells cycle slowly or arrest in the cycle) and stem cell populations expand exponentially. Because cells do not slide relative to each other in roots, relative cell lengths may be used to determine relative cell cycle durations and/or proportions of cells actively dividing in root tissues. If all cells are cycling, no interphase cells should be longer than critical length (length of longest mitotic cell in the meristem) and cells should exhibit an exponential cell – age distribution. Lengths of all cells were obtained radially across entire median longitudinal root sections at 0.5, 1.0, 1.5, 2.0, 2.5 and 3.0 mm from the founder cell/root cap boundary for five plant species to estimate percentages of cells longer than critical length. For example, up to 15 and 90% of all interphase cells were longer than critical length in 0.5 and 2.0 mm tissues, respectively, indicating that slow-cycling and/or non-proliferative cells are present in such tissues. In order to determine if the distribution of cell lengths in 0.5 mm segments approximated an exponential cell – age distribution, lengths of interphase cells less than critical length were determined. Such interphase cells were placed into ten groups according to cell length and percentages of cells in each group were compared with percentages of cells in groups calculated from an exponential cell – age distribution. Percentages of cells were significantly different from predicted percentages of between 6 and 9 out of ten groups — cell lengths were not distributed exponentially. Because there are significant numbers of interphase cells longer than critical length and since lengths of interphase cells shorter than critical length do not resemble an exponential cell – age distribution, it must be concluded that not all cells in root segments from 0.5 to 3.0 mm root segments are actively dividing. Heretofore, no databases of cell lengths have been used to test these assertions. © 2000 Elsevier Science B.V. All rights reserved.
Keywords:Root meristems; Cell lengths; Critical cell length; Exponential cell – age distribution
www.elsevier.com/locate/envexpbot
1. Introduction
Plant root meristems are complex tissues com-posed of many types of cells that have many different functions (Rost et al., 1998). Researchers
E-mail address:[email protected] (L.S. Evans)
L.S.E6ans/En6ironmental and Experimental Botany43 (2000) 239 – 251 240
have attempted to simplify the linear growth of roots (Erickson and Sax, 1956) and to minimize the variability of the types of cells that compose the terminal root area (Webster, 1980). Mathe-matical models and functions have helped re-searchers understand cell cycle kinetics and how roots grow (Erickson and Sax, 1956; Webster and MacLeod, 1980; Bertaud et al., 1986; Ivanov and Dubrovsky, 1997). By their nature, mathematical models and functions simplify otherwise complex processes. Over-simplification of otherwise com-plex processes may lead to false conclusions. This paper addresses the conclusion that all cells of root terminals are proliferative (Webster and MacLeod, 1980; Ivanov and Dubrovsky, 1997). Our approach used data of cell lengths from six locations from terminal root portions from five plant species to determine percentages of cells greater than critical length (the length of the longest cell in mitosis in the terminal portion) and the distribution of cell lengths that are shorter than critical length. If all cells shorter than critical length are proliferative, then they should approxi-mate an exponential cell-age distribution. To date, data of cell lengths in roots have not been used to address these issues.
If the assertion that all cells are dividing at the same rate so there are no slow cycling or non-cy-cling cells (Webster and MacLeod, 1980; Ivanov and Dubrovsky, 1997) is true and if cells do not slide relative to their neighbors (Sinnot and Bloch, 1939; Brumfield, 1942) then all cells of root meris-tems should have a maximum cell length equal to cells in prophase. Green (1976), Webster and MacLeod (1980) argued that if cells become non-proliferative and miss n cycles, such cells should be 2n times longer than cells which remain prolif-erative at an equivalent position along roots. Ivanov (1971) described a critical cell length for mitosis in roots of Zea mays, in which he stated that practically all cells in a tissue enter mitosis upon reaching ‘critical length’. Brumfield (1942) also cited research by Abele (1936) that supported the idea of a critical length. From all the above information, lengths of cells can be used to deter-mine relative numbers of cell cycles among neigh-boring cells. To date, few data with only limited numbers of cells have been used to determine
diversities of cell lengths within terminal root portions. Therefore, the first hypothesis addressed herein was: because all cells of root meristems are proliferative and because all proliferative cells are shorter than critical length in root meristems, there are no interphase cells longer than critical length.
Webster and MacLeod (1980), Ivanov and Du-brovsky (1997) concluded that all cells in root meristems are dividing and that the population of proliferative cells is expanding exponentially. Since the cell population is expanding exponen-tially, it should have an exponential cell – age dis-tribution in which there are two cells beginning interphase for every cell terminating interphase. In addition, there should be an exponential decay formula for cells between these two stages. Al-though an exponential condition has been as-sumed (Webster and MacLeod, 1980; Ivanov and Dubrovsky, 1997), few databases from terminal root portions have been compared with an expo-nential cell – age distribution. The second aspect assumed that interphase cells with less than criti-cal length in terminal root portions are all prolif-erative and that cell populations are growing exponentially (Webster and MacLeod, 1980; Ivanov and Dubrovsky, 1997). In this manner, the second hypothesis was: cells in terminal root por-tions that are less than critical length will have an exponential cell – age distribution based upon cell lengths.
2. Materials and methods
2.1. Microscope slides
All microscope slides used in this study were purchased from Triarch Incorporated (Ripon, WI) in 1998. This procedure of using commer-cially prepared microscope slides is not novel. Webster (1980) used microslides from a commer-cial source for his study. Ten microscope slides of median longitudinal sections of terminal root por-tions each of five plant species (Pisum sati6um
(only eight microscope slides), Pyrus communis,
Triticum aesti6um, Vicia faba, andZ.mays) were
median longitudinal root sections of P. sati6um
specifically stained to highlight chromosomes were purchased to insure that cells in mitosis for
P. sati6um would be evaluated most effectively.
The procedures to produce the excellently pre-pared microslides were provided by Dr Paul L. Conant of Triarch (personal communication). All species (P. sati6um, P. communis, T. sati6um, V.
faba and Z. mays) were commercial materials purchased from Burpee Seed Co. Roots from P.
communis were obtained from plants grown in aquaculture while roots of all other species were obtained from germinating seedlings. Seeds of these latter species were germinated on wet filter paper. All plants were grown at 21°C. Selected roots were cut from plants and fixed immediately in FAA or Navashin’s mixture (Sass, 1958). Roots were dehydrated using an ethanol – normal butanol series and transferred to paraffin. Tissues of P. sati6um, P. communis, T. aesti6um, V. faba
andZ.mays were sectioned at 10, 10, 8, 8 and 12
mm, respectively. Root sections were stained with
a quadruple stain of safranin, crystal violet, fast green and orange G.
2.2. Determination of cell lengths
Lengths of individual cells were evaluated un-der 1000 times magnification. A scale in one microscope ocular, in which one ocular unit was equal to 1.50 mm at the magnification used, was
used to determine cell lengths. To determine cell locations relative to the root terminus, all dis-tances were determined relative to the founder cell/root cap boundary. Lengths of individual cells were determined in roots at distances of 0.5, 1.0, 1.5, 2.0, 2.5 and 3.0 mm from the founder cell/
root cap boundary. Not every median longitudi-nal section had tissues at all distances. Cell lengths (lengths parallel with the long axis of the root) of all cells from the epidermis on one side of a median section radially across the entire section to the epidermis on the opposite side of a section were evaluated. Thus, only one cell from each cell file was evaluated at each distance described above. During evaluations, cell lengths were de-termined in ocular units and later converted to micrometer units. Cells were categorized as either
in mitosis (prophase, metaphase, anaphase, or telophase) or in interphase. There was no attempt to determine the phase (prophase, metaphase, anaphase, or telophase) of individual mitotic cells.
2.3. Lengths of mitotic and interphase cells
To determine percentages of cells longer than critical length, data from each segment (e.g. 0.5 mm) of all roots of a species were pooled for analysis. From these pooled data, the length of the longest cell in mitosis from all 0.5-mm seg-ments of a species was deemed critical length. For data of mitotic cells, minimum, maximum, mean and standard deviation of cell lengths were determined.
In accordance with views of Green (1976), Webster and MacLeod (1980), if a cell becomes non-proliferative and misses n cycles, it should be 2n times longer than cells which have remained proliferative at an equivalent position along the axis. Data from each segment from all roots of a species were pooled since there were no statisti-cally significant differences in cell lengths among segments. First, minimum, maximum, mean, and standard deviation of lengths of all interphase cells were determined. Second, lengths of inter-phase cells relative to critical length were deter-mined. For this step, percentages of mitotic and interphase cells in each of five categories were determined separately:
1. Cells less than or equal to critical length. Such cells may not have missed a cell division. 2. Cells longer than critical length but less than
or equal to twice critical length. Such cells may have missed one cell division.
3. Cells longer than twice critical length but less than or equal to four times critical length. Such cells may have missed two cell divisions. 4. Cells equal to or longer than four times critical length but less than or equal to six times critical length. Such cells may have missed three cell divisions.
L.S.E6ans/En6ironmental and Experimental Botany43 (2000) 239 – 251 242
2.4. Estimation of cell–age distribution 6ia cell
lengths
In accordance with views that plant root meris-tems consist of exponential cell – age distributions (Webster and MacLeod, 1980; Ivanov and Du-brovsky, 1997), and all cells in a meristem enter mitosis when they obtain critical length (Ivanov, 1971), a frequency distribution of cell lengths of cells shorter than critical length in 0.5 mm root segments will be comparable to a theoretical cell – age distribution for each species tested. Once crit-ical cell length was determined for each species, cells were grouped within each 1.5-mm length
be-tween minimum and critical lengths determined for each root segment. A calculated cell – age dis-tribution (e.g., 10, 20%, etc. of the cycle com-pleted) was calculated with cell length data. For all root segments, there were a small number of cells that were very short. For all segments, these small numbers of very short cells were grouped with longer cells in order to obtain a more realis-tic value for the first decile (0 – 10%) of the cell – age distribution. After the number of cells of the first decile was established, cell numbers in re-maining deciles were determined by partitioning the remaining cells in roughly equal cell length groupings. In this manner, numbers of cells in ten decile groups were established for each segment. From numbers of cells in deciles, percentages of cells in deciles of the cycle were calculated. Calcu-lated percentages of cells in each decile of cycle completed were compared with expected percent-ages of cells in deciles based upon an exponential decay formula yt=y0−kt, where yt and y0 are
relative frequencies at time t and time 0, respec-tively (Webster and MacLeod, 1980). Expected percentages of cells in each decile of ten root samples at each tissue distance were compared with the expected percentage with a chi-square test (Mendenhall and Ott, 1972).
3. Results
3.1. Lengths of mitotic and interphase cells
Distributions of cell lengths of cells in mitosis
from all ten, 0.5-mm segments ofP.sati6umwere
determined. Data in Fig. 1 show distributions of cells in mitosis for three separate, 0.5-mm root samples (three roots with the largest numbers of cells in mitosis). Each of the three figures show two peaks which may be attributed to smaller cells which may have been in telophase and larger cells which may have been in prophase, metaphase and anaphase. Clearly, there was no distinct separation between smaller and larger mi-totic cells within any of the three roots. Mean length values ranged from 25.0 to 32.6 mm and
standard deviations ranged from 8.2 to 10.4 mm.
Data in Table 1 show a composite distribution of all mitotic cells from all 0.5-mm segments of P.
sati6um tested. Lengths ranged from 9 to 56 mm
and showed no distinct separation between smaller and larger mitotic cells with a mean of 27.4 mm and a standard deviation of 9.2 mm.
Overall, there were 24 cells (11.5%) above 37.5
mm in length with no values greater than 52 mm
(critical length).
Data in Table 1 also show that mitotic cells of
P. sati6um were longer in 1.0 – 3.0 mm segments
compared with 0.5-mm segments. For example, for 1.0-mm segments, mean length was 25.6 mm,
similar to 27.4 mm for 0.5-mm segments, while
for cells from 2.5-mm segments mean length was 49.9 mm. In accordance, 5.5% of all mitotic cells
had lengths greater than 52 mm in 1.0-mm
seg-ments while 33.3% of mitotic cells in 2.5-mm segments were greater than 52 mm. In this
man-ner, mitotic cells were larger in root tissues far-ther from terminal portions.
The distribution of lengths of interphase cells in 0.5-mm segments of P. sati6um was compared
with distributions of lengths of mitotic cells shown above in order to determine numbers and percentages of cells longer than critical length. Lengths of interphase cells of 0.5-mm root seg-ments of P. sati6um ranged from 7.5 to 135 mm
and 8.7% of all cells had lengths greater than critical length (Table 1). As noted for mitotic cells above, interphase cells of P. sati6um were larger
L
.
S
.
E
6
ans
/
En
6
ironmental
and
Experimental
Botany
43
(2000)
239
–
251
244
Table 1
Cell length characteristics in various segments in roots of five plant species
Percentage of cells in relation to critical lengths
STATISTICAL PARAMETERS Number of Type of Cells
Species (Criti- Distance Minimum Maximum Mean Length Standard De- Below One Between One Between Two Between Four More than and Two
viation (mm)
Cells Critical and Four and Six
Criti-cal Length) From Root Length (mm) Length (mm) (mm) Six Critical
Critical
Critical Lengths
Length cal Lengths
cap (mm)
Lengths Lengths
214 9.00 52.0 27.4 9.2 100 0 0 0 0
Mitosis
Pisum sati6um 0.5 (52.5mm)
25.6 12.8 94.5 5.5 0
7.5 0
219 90.0 0
Mitosis 1.0
26.0 9.0 100.0 0 0 0
Mitosis 1.5 59 13.5 51.0
44.6 25.0 77.8 18.5 0 3.7
100 0
16.5 27
2.0 Mitosis
24.0 94.5 49.9 20.5 66.7 33.3 0 0 0
2.5
Mitosis 18
64.5 20.4 25.0 75.0 0 0
90.0 0
Mitosis 3.0 12 27.0
27.2 18.7 91.3 7.5 1.2 0
Interphase 0.5 509 7.5 135 0
38.1 29.9 79.8 15.4 4.8 0
195 0
7.5 524
1.0 Interphase
68.5 57.8 53.6 24.3 18.1 3.8 0.2
Interphase 1.5 439 9.0 420
103 82.2 31.7 29.8 26.4 11.2
559 0.9
15.0 Interphase 2.0 420
117 83.1 19.3 34.1 32.1 13.8 0.7
Interphase 2.5 405 15.0 510
135 75.8 8.9 27.9 45.4 17.2 0.6
phase cells were longer than 105 mm compared
with 46.7% in 2.5-mm segments. Thus, there was a marked increase in cell elongation in mature root tissues.
Lengths of cells in mitosis in 0.5-mm segments of Pyrus communis are shown in Table 1. In 1.0 – 3.0 mm segments of P. communis, only nine cells were in mitosis. In 0.5-mm root segments of
P. communis, 8.1% of all interphase cells were longer than critical length (33 mm). In contrast,
more than 50% of all interphase cells were longer than critical length in 1.5-mm segments. More-over, a few cells in 1.5-mm segments were longer than eight times critical length.
Only root segments of 0.5 and 1.0 mm were present in roots ofT.aesti6um. Lengths of cells in
mitosis of 0.5 mm segments provided a critical length of 37.5mm. Few mitotic cells were found in
0.5 mm segments and no mitotic cells were found in root segments at 1.0 mm (Table 1). In 0.5 and 1.0 mm root segments of T. aesti6um, 1.8 and
57.4% of all interphase cells were longer than critical length, respectively. In 1.0 mm segments, 19% of all interphase cells were longer than four times critical length.
Critical length of mitotic cells in 0.5 mm seg-ments of V. faba was 48 mm (Table 1). Mean
lengths of mitotic cells were similar for 0.5 – 1.5 mm segments (27.7 – 36.3 mm). Few mitotic cells
were present in 2.0 – 3.0 mm segments. In 0.5 mm segments of V.faba, 15.5% of all interphase cells were longer than critical length and 6.4% of all interphase cells were longer than four times criti-cal length. Twenty-five, 36, 42, 60 and 69% of all interphase cells were longer than critical length at 1.0, 1.5, 2.0, 2.5, 3.0 mm, respectively.
Few cells in mitosis were present in 0.5 and 1.0 mm root segments ofZ. mays(Table 1). No cells in mitosis were present in 1.5 – 3.0 mm segments. Among interphase cells ofZ.mays, 8.1 and 27.0% of interphase cells were longer than critical length (30 mm) at 0.5 and 1.0 mm, respectively. In 2.5
and 3.0 mm segments, 5.4 and 3.8% of all inter-phase cells were less than critical length. Lengths of interphase cells varied markedly in 3.0 mm segments in which 22.1, 40.4, 29.8 and 3.9% were longer than 30, 60, 120 and 240mm, respectively.
To test the first hypothesis, namely no
inter-phase cells in root meristems will be longer than critical length, cells in root segments of 0.5 – 3.0 mm from the root cap-founder cell boundary were subdivided by cell length. Cells were subdivided into length groups relative to critical length of mitotic cells of 0.5-mm segments. In general, data of all five plant species demonstrate that from 7 to 10% of all interphase cells were longer than criti-cal length in 0.5 mm root segments. Moreover, larger percentages (up to 57%) of cells were longer than critical length for 1.0 mm root segments of
T. aesti6um. If a root meristem region is
consid-ered to be contained in 0.5 – 3.0 mm root segments of the five species tested, the first hypothesis is not supported.
3.2. Estimation of cell–age distributions
A frequency distribution of cells with cell lengths less than critical length (56 mm) for P.
sati6um demonstrate that in almost all
compari-sons there was a statistically significant difference between actual percentages compared with theo-retical values. Percentages of larger cells of P.
sati6um were below expected percentages.
For P. communis, cell lengths considered for this analysis ranged from 7.5 to 34.5 mm (Table
2). For statistical analysis, almost all decile groups contained cell lengths of 3 mm except cells of 15
and 16.5 and 16.5 and 18mm (Table 2). These two
groups were considered as separate groups be-cause they each had large percentages of cells. Analyses shown in Table 2 indicate better corre-spondence between actual and theoretical data for shorter cells than for longer cells. For example, the three groupings of longest cells had 13.2% of all cells compared with a theoretical value of 23.1%. The relatively high percentages of 16.9 and 16.4% for deciles of 21 – 30 and 31 – 40 were sig-nificantly above theoretical values of 11.7 and 10.1%, respectively.
Lengths of cells of T. aesti6um did not appear
to be similar to a theoretical exponential cell – age distribution (Table 2). All decile groupings con-tained cells of at least a 3-mm range (Table 2).
L
Results of a Chi-square analysis of percentage of cells based upon cell lengths in individual roots compared with percentages based upon an exponential cell-age distribution in deciles of the cycle completed
Percentages of cells in each decile (Probability)
Species or Theoretical Distribution Decile
21–30 31–40 41–50 51–60 61–70 71–80 81–90 91–100
0–10 11–20
(0.01) (0.05) (0.01) (0.05) (0.01) (0.10)
(0.01) (0.01)
(0.005) (0.005)
16.9 16.4 10.9 8.3 9.6 4.8 1.9 6.5
14.6
Pyrus communis 10.1
(0.01) (0.01) (0.10) (0.10) (0.10) (0.05) (0.005) (0.10)
(0.10) (0.05)
13.5 8.3 8.8 7.9 4.6 2.3
23.3 0.5
Triticum aeti6um 15.4 15.4
(0.05) (0.005) (0.05) (0.05) (0.10) 10.10) (0.05) (0.01) (0.005)
(0.05)
17.6 10.4 16.1 6.3 4.8 1.5
8.7 3.9
Vicia faba 16.7 14.0
(0.05) (0.01) (0.10) (0.005) (0.05) (0.05) (0.005) (0.01)
(0.01) (0.05)
9.2 11.8 26.5 13.2 5.2 8.0 4.0 4.8
8.8 8.5
Zea mays
(0.10) (0.005) (0.05) (0.01) (0.10) (0.01)
(0.05) (0.01)
(0.01) (0.005)
33.0
Mitosis 0.5 26 13.5 22.2 6.3 100 0 0 0 0
Pyrus communis(33.0mm)
29.0 24.0 75 12.5 12.5 0
97.5 0
10.5
Mitosis 1.0 16
40.5 385 75 25.0 0 0 0
Mitosis 1.5 4 15.0 97.5
37.5 0 0 100 0 0
Interphase 1.0 403 7.5 180 25.8 24.4 76.2 16.4 6.4 1.0 0
50.3 47.7 49.4 27.0 17.1 5.7
Interphase 2.0 387 15 124 140 8.5 34.1 31.5 13.7 12.2
103 95 2.0 18.5 41.9 23.0
907 14.6
22.5
Interphase 2.5 356
191 156 0.6 7.4 41.4 28.1 22.5
Interphase 3.0 324 35 601
21.6 7.5 100 0 0 0
10.5 37.5 0
16 0.5 Mitosis
TriticumAestivum (37.5mm)
– – – – – – –
Mitosis 1.0 0 – –
17.6 9.4 98.2 0.9 0.9 0
9
Interphase 0.5 219 90
112 46.9 25.0 42.6 38.3 19.1 0 0
Mitosis 2.5 2 63 142
.
E
6
ans
/
En
6
ironmental
and
Experimental
Botany
43
(2000)
239
–
251
247
Table 2
Percentages of cells in each decile (Probability)
Species or Theoretical Distribution Decile
31–40 41–50 51–60 61–70 71–80 81–90 91–100
0–10 11–20 21–30
35.4 31 84.5 8.6 6.4 0.5
7.5 210 0
394 0.5 Interphase
Interphase 1.0 387 7.5 450 39.9 45.6 74.9 17.0 7.2 0.3 0.6
Interphase 1.5 388 7.5 300 52.6 53.8 64.1 20.1 11.9 3.9 0
65.9 59.0 57.8 20.1 17.1 5.0
375 0
341 10.5
Interphase 2.0
345
Interphase 2.5 209 15 77.6 65.0 39.7 37.3 16.8 6.2 0
308
Interphase 3.0 120 18 90.8 63.3 30.8 36.7 25.0 7.5 0
22.1 5.7 100 0 0 0
18 30 0
4 0.5 Mitosis
Zea mays(30.0mm)
20.2 13.6 75 25 0 0 0
Mitosis 1.0 8 7.5 48
– – – – – –
– –
Mitosis 1.5 0 –
–
Mitosis 2.0 0 – – – – – – – –
– – – – – – –
Mitosis 2.5 0 – –
– – – – – – –
–
Mitosis 3.0 0 –
Interphase 0.5 418 6 90 17.0 9.8 91.9 6.7 1.4 0 0
150
Interphase 1.0 411 6 24.3 20.7 73.0 20.2 5.8 1.0 0
54.9 61.0 36.0 32.2 22.8 6.4
450 2.6
311 6
Interphase 1.5
450
Interphase 2.0 161 15 106 88.0 8.1 27.3 31.1 24.2 9.3
131
Interphase 2.5 148 12 480 101 5.4 20.3 29.0 30.4 14.9
103 64.6 3.8 22.1 40.4 29.8 3.9
375
L.S.E6ans/En6ironmental and Experimental Botany43 (2000) 239 – 251 248
distributions differed significantly from the theo-retical distribution. Most particularly, percentages of larger cells were below expectations.
Lengths of cells of V. faba did not resemble a theoretical cell – age distribution (Table 2). Actual percentages (17.6 and 16.1%) were not similar to theoretical percentages (10.9 and 9.5%, respec-tively) in 31 – 40 and 51 – 60 deciles. Moreover, actual percentages (4.8, 1.5, and 3.9%) for the three deciles of longest cells were below theoreti-cal percentages (8.2, 7.7 and 7.2%, respectively). Both percentages of smallest and largest cells were below expectations.
Lengths of cells ofZ. mayswere not similar to a theoretical exponential cell – age distribution (Table 2). The actual value of 26.5% for decile 41 – 50 was almost three times larger than the theoretical percentage (10.1%).
To test the second hypothesis, namely — inter-phase cells less than critical length in root meris-tems will have an exponential cell – age distribution, distributions of lengths of cells of 0.5 mm root segments that were less than critical length were compared with cell lengths based upon a theoretical exponential cell – age distribu-tion. As stated above, for a theoretical exponen-tial cell – age distribution there should be two cells at the beginning of a cell cycle for every one cell at the end of a cell cycle and the number of cells throughout the cycle should decrease exponen-tially. Results demonstrate that actual cell length distributions of 0.5 mm root segments of five plant species were not similar to such theoretical distributions. Overall, percentages of the smallest cells exceeded expectations while percentages of largest cells were below expectations. An expo-nential cell – age distribution is not supported for 0.5-mm segments of five plant species.
4. Discussion
There is a concept that root terminals contain founder and initial cells which produce cells for terminal root meristems (Barlow, 1976). Founder cells produce initial cells when necessary. First generation initial cells undergo cell divisions to produce second generation initial cells, which
re-main in the same relative location as their parents as well as first generation derivative cells. First generation derivative cells are displaced basipetally relative to second generation initial cells and may begin the process of producing derivative cells (Luck et al. 1994). According to Barlow (1976), derivative cells constitute the ma-jority of the volume of terminal root portions of roots (root meristems). Cell divisions of derivative cells may produce an exponential increase in cell number (e.g., 2, 4, 8, 16, 32 cells). This possible exponential increase in cell number has prompted some researchers (Webster and MacLeod, 1980; Ivanov and Dubrovsky, 1997) to conclude that virtually all cells in root meristems are prolifera-tive (a growth fraction of 100%). It is possible that although many cells may conform to the above pattern, some cells may cease dividing and elongate relative to their shorter, cycling neighbors.
Although it has been concluded that growth fractions (percentage of cells that divide) of root meristems are virtually 100% (Webster and MacLeod, 1980; Ivanov and Dubrovsky, 1997), little or no scientific evidence has been forthcom-ing to support this conclusion. Other researchers believe that not all cells in terminal root segments are proliferative. For example, Clowes (1971) cal-culated growth fraction percentages by several different methods for four cell types in roots of
Zea mays. Of several cell types measured, only stele cells at 200 mm from the quiescent center/
root cap boundary had a growth fraction at or near 100%. In other cell types, estimates of up to 20 – 25% non-cycling cells were common. The re-sults of Burholt and Van’t Hof (1971) are ger-mane. These researchers found that root growth of Helianthus annuus was linear for the tempera-ture range of between 10 and 30°, the total num-ber of cell files and mitotic index did not vary among roots grown at various temperatures, and the growth fraction varied from 0.75 to 0.93. Moreover, Evans and Van’t Hof (1975) estimated growth fractions of between 0.47 and 0.89 for 0 – 2-mm terminal root segments of four plant species.
from estimating numbers of cell divisions each cell completes. If numbers of division cycles were identical and cell elongation rates were identical then all cells would be displaced basipetally at the same rate. This condition does not appear to operate in terminal root segments. Barlow (1976) demonstrated that numbers of completed cycles varied among root tissues. For example, xylem cells of Z. mays were said to have two to four division cycles before cell mat-uration while cortical cells had seven division cycles before maturation. If sliding growth does not occur (Sinnot and Bloch, 1939; Brumfield, 1942), then heterogeneity of cell lengths demon-strates differences in numbers of cycling cells and/or rates that cells cycle. Bertaud et al. (1986) found significant variations in lengths of cells in terminal 72 to 263mm root tissues of Z.
mays in which 1.0, 10.3, 28.9, 32.0, 20.6, 5.2 and 2.0% of all cells in these segments were between 9, 9.1 – 14.0, 14.1 – 19.0, 19.1 – 24.0, 24.1 – 29.0, 29.1 – 34.0 and 34.1 – 39.0 mm in length,
respec-tively. Results reported herein support those of Barlow (1976), Bertaud et al. (1986) and point to heterogeneity of cell lengths and thus in dura-tions of cell cycles.
Data of Z. mays presented by Ivanov (1971) are difficult to parallel with results mentioned above of Clowes (1976), Barlow (1976), Bertaud et al. (1986), and the results of the present study. Ivanov (1971) measured between 140 and 191 cells (on average) per file for roots of Z.
mays. If we assume a mean cell length of 17 mm
(mean cell length of cells in 0.5 mm segments from the present study) then Ivanov (1971) would have measured 140 cells or 2.38 mm dis-tance (140 cells times 17 mm per cell) from the
root cap/quiescent cell boundary. It is unclear why Ivanov did not experience any cells longer than 40 mm (Fig. 4 of Ivanov, 1971). In
con-trast, Clowes (1976) demonstrated that cell lengths at 300 mm behind the quiescent center
varied by a factor of three. In the stele, a factor of 11 was reached at 1.2 mm of the root (Clowes, 1976). In the present study, 24.2% of all interphase cells were between 120 and 240
mm and 9.3% of all interphase cells were more
than 240 mm at 2 mm from the root cap/
quies-cent quies-center boundary. Moreover, the model de-veloped by Bertaud et al. (1986) which used data from Erickson and Sax (1956) demon-strated sharp differences in cell production rates between cortical and stellar tissues (see Table 11 of Bertaud et al., 1986). This model demon-strated non-proliferative cells from 375 and 625
mm onward in the stele and cortex, respectively.
There seems to be substantial differences be-tween observations of Ivanov and three other studies (Clowes, 1976; Bertaud et al., 1986;present study).
Brumfield (1942) in his classic study with growing roots of Phleum pratenseconcluded that such roots had three phases, a (1) new proto-plasm phase which occurred from 0 to 140, a transition phase from 140 to 310, and a vacuole enlargement phase from 310 to 970 mm. Results
of Goodwin and Stepka (1945) for P. pratense
are noteworthy. Roots of this species exhibited cells from 60 to 260 mm at 1.0 mm from the
terminal area. The results of these two studies with P. pratense appear coincident with the re-sults of Clowes (1976), Bertaud et al. (1986) and the present study with Z. mays in the respect that cell arrest occurs rather close to the termi-nal area.
Webster (1980) demonstrated that lengths of mitotic cells increased at greater distances from root terminals in P. sati6um and Hyacinthus ori
-entalis. Considerable heterogeneity in cell length among cell types from shorter inner cortex cells to longer outer stele cells was present in both species. For P. sati6um, outer stele mitotic cells
were up to 60 mm in length while inner cortex
cells were 15 mm in length at 1.0 mm from
the terminal. Likewise, outer stele mitotic cells were more than 120 mm long while epidermal
cells were 25 mm in length at 2.0 mm for H.
orientalis. These values are not unlike values re-ported herein. The wide diversity of mitotic cells documented above was small relative to diversi-ties of lengths of interphase cells at 2.0 and 3.0 mm distances. Although many species such as P.
sati6um, and V. faba had less than one percent
of cells longer than six times critical length in 3.0 mm segments, roots of P. communis and Z.
L.S.E6ans/En6ironmental and Experimental Botany43 (2000) 239 – 251 250
than six times critical length in more mature root tissues. These large diversities in lengths reflect diversities in cell cycle durations (Clowes, 1976). Seemingly, this complex tissue with its many, varied cell types is not character-ized easily and should not be reduced to mathe-matical expressions without realizing the complex-ity of the tissue.
It may be stated that the analysis performed herein should be performed with each tissue indi-vidually and that combining data from all tissues is inappropriate. The above idea neglects several aspects. To understand how a root meristem works it must be considered as a whole unit with various tissues. First, the root meristem acts clearly as a unit that contains a mixture of tissues. To really understand how a root meristem works it must be taken as a whole unit with various tissues. The various tissues appear to conform to the structure of the meristem. Second, the proce-dure of evaluating terminal roots as a whole has a long-standing history in the botanical literature. The classical paper by Erickson and Sax (1956) used all tissues from roots of Z. mays. Green’s (1976) paper dealt with growth and cell patterns in terminal roots considered all root tissues. Of course, some papers have considered kinetics of individual tissues (Clowes, 1971, 1976) but in the main, most researchers have pooled data from all tissues of the meristem.
Criteria that define a plant root meristem have never been established precisely so that one cell in a file is considered in the meristem while its neighbor is not in the meristem. In other words, the demarcations of the limits of the root meris-tem have not been defined. For example, it has been shown by many researchers (noted above), that cells in some files (tissues) in roots stop cell division before cells in other files (other tissues) [see many examples above]. How does the concept of a root meristem relate to, for example, two files of cells (possibly from two separate tissues) in which actively dividing cells occur in one file while non-dividing, elongating cells occur in the adja-cent file? If the researcher acknowledges that the above situation occurs in terminal root portions, (as shown by the results of this paper) then only one of the following three categories is correct for
the two cells considered above: 1. both cells are ‘in’ the meristem. 2. both cells are ‘out’ of the meristem
3. the meristem is a discontinuous, amorphous grouping of cells in which dividing cells in one file are ‘in’ the meristem while non-dividing cells in an adjacent file are ‘out’ of the meristem.
If alternative ‘a’ is chosen then root meristems are composed of several types of cells including cells with rapid, continuous cell divisions, cells that undergo cell divisions less frequently, and possibly cells that have stopped progress in the cell cycle. If alternative ‘b’ is chosen, then root meristems are composed of only rapidly cycling cells. All such cells would have cell lengths shorter than cells in prophase. Such meristems would be very small because if a small cell is located adja-cent to a long cell, then one would have some cell files with large numbers of cells while other cell files may have only relatively few cells. In short, if alternative ‘b’ is chosen, the most important crite-rion is the presence of short cell cycles coupled with little cell elongation.
References
Abele, K., 1936. Zur kennetis der zell-und kernteilung in dem primaren meristem. Protoplasma 25, 92 – 114.
Barlow, P.W., 1976. Towards an understanding of the behav-ior of root meristems. J. Theor. Biol. 56, 433 – 452. Bertaud, D.S., Gandar, P.W., Erickson, R.O., Olliver, A.M.,
1986. A simulation model for cell growth and proliferation in root apices. I. Structure of model and comparisons with observed data. Ann. Bot. 58, 285 – 301.
Brumfield, R.T., 1942. Cell growth and division in living root meristems. Am. J. Bot. 29, 533 – 543.
Burholt, D.R., Van’t Hof, J., 1971. Quantitative thermal-in-duced changes in growth and cell population kinetics of Helianthusroots. Am. J. Bot. 58, 386 – 393.
Clowes, F.A.L., 1971. The proportion of cells that divide in root meristems ofZea mays. Ann. Bot. 35, 249 – 261. Clowes, F.A.L., 1976. Estimation of growth fractions in
meris-tems ofZea maysL. Ann. Bot. 40, 933 – 938.
Erickson, R.O., Sax, K., 1956. Rates of cell division and cell elongation in the growth of the primary root ofZea mays. Proc. Am. Philos. Soc. 100, 499 – 514.
Evans, L.S., Van’t Hof, J., 1975. The age-distribution of cell cycle populations in plant root meristems. Exp. Cell Res. 90, 401 – 410.
Goodwin, R.H., Stepka, W., 1945. Growth and differentiation in the root tip ofPhleum pratense. Am. J. Bot. 32, 36 – 46. Green, P.B., 1976. Growth and cell pattern formation on an
axis: critique of concepts, terminology, and modes of study. Bot. Gaz. 137, 187 – 202.
Ivanov, V.B., 1971. Critical cell size and transition to cell division. I. Sequence of transition to mitosis of sister cells in the root tip of maize seedlings. Ontogenez 2, 524 – 535. Ivanov, V.B., Dubrovsky, J., 1997. Estimation of the cell-cycle duration in the root apical meristem: a model of linkage between cell-cycle duration, rate of cell production, and rate of root growth. Int. J. Plant Sci. 158, 757 – 763. Luck, J., Barlow, P., Luck, H., 1994. Deterministic patterns of
cellular growth and division within a meristem. Ann. Bot. 73, 1 – 11.
Mendenhall, W., Ott, L., 1972. Understanding Statistics. Duxbury Press, Belmont, CA, p. 310.
Rost, T.L., Barbour, M.G., Stocking, C.R., Murphy, T.M., 1998. Plant Biology. Wadsworth Publishing Co., New York.
Sass, J.E., 1958. Botanical Microtechnique, 3rd edn. Iowa State University Press, Ames.
Sinnot, E.W., Bloch, R., 1939. Changes in intercellular rela-tionships during the growth and differentiation in living plant tissues. Am. J. Bot. 26, 625 – 634.
Webster, P.L., 1980. Analysis of heterogeneity of relative division rates in root apical meristems. Bot. Gaz. 141, 353 – 359.
Webster, P.L., MacLeod, R.D., 1980. Characteristics of root apical meristem cells population kinetics: a review of anal-ysis and concepts. Environ. Exp. Bot. 20, 335 – 358.
