J. Cell Sci. 69, 87-105 (1984)
87
Printed in Great Britain © The Company of Biologists Limited 1984
SCINTILLATION AND AUTORADIOGRAPHIC
STUDIES ON 63NICKEL UPTAKE IN PSEUDOMONAS
TABACI
R. H. AL-RABAEE AND D. C. SIGEE*
Cytology Unit, Departments of Botany and Zoology, University of Manchester,
Manchester, U.K.
SUMMARY
Scintillation studies on the uptake of 63 Ni 2+ by Pseudomonas tabaci demonstrate an incorporation
of approximately 2-5 nmol per 10'° bacterial cells, in medium containing 12nmol (8-3 /iCi) per ml.
Over 80% of the incorporated Ni 2+ is lost from the cells during washing, fixation and dehydration
with ethanol. The remaining insoluble (bound) 63 Ni 2+ has the highest level in cells fixed in acetic
acid/ethanol (0-4 nmol/10 10 cells), with smaller amounts in paraformaldehyde- and glutaraldehydefixed cells. The radioactive level in aldehyde-fixed cells represents a total Ni 2+ uptake of about
lCTl8g or 104 atoms per cell.
Light- and electron-microscope autoradiography corroborated the scintillation studies in demonstrating a higher retention of label by cells fixed in acetic acid/ethanol, possibly reflecting a higher
retention of medium M, proteins with this type of fixation. High-resolution electron-microscope
autoradiography involving gold latensification with physical development demonstrated a clear
localization of silver grains to the central nucleoid region (seen most clearly over the discrete
nucleoid of aldehyde-fixed cells) and within this to the chromatin (seen most clearly over the
condensed chromatin of acetic acid/ethanol-fixed cells). It is suggested that the incorporated
63
Ni 2+ labels mainly central, genetically inactive DNA, while peripheral, actively transcribing DNA
has little associated radioactivity. The pattern of cation association seen in this bacterium shows a
number of close similarities to the situation seen in dinoflagellate cells.
INTRODUCTION
The uptake of nickel from the environment, and its biological effects on whole
organisms, tissues and cells, has increasingly occupied the attentions of research
workers in recent years (Sunderman, 1978; Brown & Sunderman, 1980; Hutchinson,
1981). Parallel to this development has been the need for improved methods of nickel
detection in biological and environmental samples (Dulka & Risby, 1976; Sunderman, 1980a,b; Jaworski, 1974), with particular emphasis being given to the determination of nickel levels in liquids — including water samples, body fluids, tissue and
cell homogenates and solutions. This chemical analytical approach gives an overall
picture of nickel levels, but provides little information about the location of nickel
within tissues and within individual cells. Various techniques are available, however,
at the levels of the light and electron microscopes for intracellular nickel quantitation,
including X-ray microanalysis (Kearns & Sigee, 1980; Sigee & Kearns, 1980) and
63
Ni2+ autoradiography (Sigee, 1982).
• Requests for reprints should be sent to this author.
88
R. H. Al-Rabaee and D. C. Sigee
The purpose of the work reported here was to investigate the uptake of 63Ni2+ into
bacterial cells cultured in the laboratory, with particular reference to the localization
of incorporated label at the level of the electron microscope. The use of 63 Ni 2+
(Kasprzak & Sunderman, 1979) provides a highly sensitive measure for monitoring
relatively low levels of Ni2+ uptake, for which both liquid scintillation counting and
autoradiography can be used. The use of these techniques, on cells processed in a
variety of ways, gives information on the presence and location of insoluble nickel in
bacterial cells. The results obtained are of general interest in relation to the metabolic
role and toxicology of Ni2+ in bacteria. More particularly, they give direct information
on the association of Ni ions with bacterial chromatin, and in this respect have
relevance to previous studies carried out on dinoflagellate (eucaryotic) cells (Sigee,
1983a,6).
MATERIALS
AND
METHODS
Culture of bacteria
Cultures of Pseudomonas tabaci, originally obtained from Professor R. N. Goodman, University
of Missouri, were grown in nutrient broth in a shaker at 23 °C to log phase. A sample (1 ml) of
bacterial culture was added to 99 ml of nutrient broth (Oxoid), and the bacterial population was
determined every 4h by making direct counts using a haemocytometer. Atomic absorption
spectrophotometric studies on the nutrient broth showed that the level of Ni 2+ in normal growth
medium was below the limits of detection (1 part per million) of the technique.
Labelling procedure
Radioactive label was added to the bacterial culture 5 h after inoculation, at the beginning of the
logarithmic phase of growth (Fig. 1); 10ml of "nickel chloride (Amersham International Ltd) in
0-1 M-HC1 was adjusted to pH 7-2 (the pH of the bacterial culture) by addition of 0-5 M-Tris buffer,
then added to 50 ml of cell culture to give an overall concentration of 8 3 fiCi/ml (sp. act. of added
Ni 2+ , ll-6mCi/mg). Samples were taken after labelling periods of 1 —16h (Fig. 1), washed in
sodium cacodylate buffer (01 M, pH 7-2), and used for scintillation counting and autoradiography.
Cell processing
The level of incorporated Ni2"1" was determined either in fresh, unfixed cells (soluble and insoluble
cations: scintillation counting) or in fixed, chemically dehydrated cells (insoluble cations: scintillation counting and autoradiography). Fixation was carried out at 20CC using the following fixatives,
(a) Glutaraldehyde, 2-5 % in 0-1 M-sodium cacodylate buffer (pH 7-2), for 2-3 h. Some cells were
also postfixed (2h) in 2 % buffered osmium tetroxide for autoradiography. (b) Paraformaldehyde,
4 % freshly-prepared solution in 0-1 M-sodium cacodylate buffer, for 2-3 h. (c) Acetic acid/ethanol
(1:3, v/v), for 2-3 h. Cells fixed in acetic acid/ethanol were transferred directly to 100% ethanol
(Fig. 2), while aldehyde-fixed cells were dehydrated in an ethanol series (Fig. 3).
Scintillation counting
Scintillation studies were carried out both on fluids used to process the cells (supernatants) and
the final cell suspension. For each washing, fixation or dehydration stage 10 ml of fluid was added
to the cell pellet, and 1 ml of liquid taken from the resulting supernatant and used for scintillation
counting. This was added to 10 ml of Packard scintillation fluid (type 299 TM) and allowed to stand
in the dark for 48 h to eliminate errors due to chemiluminescence (Kasprzak & Sunderman, 1979).
Counts were made in a Packard liquid scintillation spectrometer, model 300CD, over an interval of
lOmin, and were converted to radioactive concentrations (/iCi/ml) by reference to a standard
calibration curve. The overall counting efficiency (c.p.m.X 100/disints per min) was 47%. Bacterial population counts (cells/ml) were made from the final samples.
63
Ni2+ uptake in Pseudomonas tabaci
89
Autoradiography
Light- and electron-microscope autoradiography was carried out on separate samples from those
used for scintillation counting. Fixed, dehydrated cells were embedded in Spurr resin and sectioned
at a thickness of Zfim (light microscopy) or 60nm (electron microscopy). Light-microscope
preparations were coated with a layer of Ilford G-5 emulsion, using a dipping technique. For the
electron microscope, ultrathin sections were coated with a monolayer of Ilford L4 emulsion using
a loop method (Williams, 1977). Preparations were incubated in a light-tight box for periods of 7
days (light microscopy) and up to 6 months (electron microscopy), and were then processed using
D19 developer and 10% Hypam fixative. Some electron-microscope preparations were also
processed by fine-grain physical developer (paraphenylenediamine) preceded by gold latensification, according to the method of Salpeter & Bachmann (1964).
RESULTS
Scintillation counts
In separate experiments, scintillation studies were carried out to determine the
effect of varying: (a) the type of cell processing (fixation experiment), and (b) the
labelling period (time-course experiment).
Fixation experiment. Cells labelled for 2h (Fig. 1) were washed three times in
buffer and then either sampled directly or fixed and dehydrated. For each processing
schedule, scintillation counts were converted to ^Ci, and expressed as radioactivity
per 1010 bacterial cells (Figs 2, 3). These data provide a direct comparison between
different treatments in the form of a 'balance sheet', from the second wash to the final
cell suspension. The cumulative totals for each schedule were constant, at about
2-1 jUCi/1010 cells, indicating that overall comparison between treatments was valid.
Supernatant from the first wash was not included since this would contain a high
proportion of original radioactive medium.
With each treatment, approximately half the 63Ni2+ contained in the cells after the
first wash was lost in the second and third washes. This represents a pool of watersoluble label that is readily lost from the cells by simple diffusion. The remaining
label, totalling about O-9^Ci/lO10 cells in the final suspension (Fig. 2A), represents
bound and soluble 63 Ni 2+ retained during the washing. Fixation in acetic acid/ethanol
and subsequent dehydration in 100 % ethanol (Fig. 2B) removes much of the soluble
63
Ni2+ by direct extraction, leaving a largely insoluble component at about 0-3 ^Ci/
1010 cells in thefinalsuspension. Fixation in paraformaldehyde (Fig. 3A) and glutaraldehyde (Fig. 3B), followed by dehydration, results in a more complete extraction of
soluble 63 Ni 2+ , leaving about 0-2^Ci/1010 cells in the final suspension. The
progressive drop in radioactive content of the processing liquids during these
dehydration series (Fig. 3A,B), with very low levels in the 90% and 100% ethanol,
suggests that very little, if any, of the extracted 63Ni2+ is lipid-associated.
Analysis of the final bacterial suspensions is presented in Table 1, for unfixed cells
in buffer and forfixedcells in 100 % ethanol. On the basis that all the label is contained
in the suspended bacterial cells, values for the mass of incorporated Ni2+ vary from
3-0x10"" to 12-3xlO~ u nmol/cell. These are equivalent to a retention of 1-8X104
to 7-4X104 Ni2+ atoms/cell. The level of nickel retained by cells fixed in acetic
CEL69
90
R. H. Al-Rabaee and D. C. Sigee
Time (h)
Fig. 1. Growth curve of P. tabaci and associated radiolabelling periods. Mean values of
population count (each derived from three separate readings made using a haemocytometer)
are given at various times after inoculation of high nutrient broth with dense culture.
acid/ethanol amounts to about 32% of that in unfixed cells, and in the case of the
aldehyde fixations the level falls to 24%.
Time-course experiment. Cells were labelled for periods of 1, 2, 8 and 16 h, commencing during the early log phase of bacterial growth (Fig. 1). Although the radioactive content of the final suspension shows a considerable increase (from 0-4x 10~3 to
12-6xlO~3;uCi/ml), the level of incorporated 63Ni2+ per bacterium (Table 2) remains
fairly constant at about 1-4X 10" " nmol/cellfrom 1-8 h, dropping to 0-8 XlO~" nmol/
a
Ni2+ uptake in Pseudomonas tabaci
91
100-8
0-6
0-2 -
1
j?
Wash
Cells
Wash
3
AA fix
1 0
I °'8
to
a:
0-4
0-2
Wash
2
Ethanol
(100%)
Sample
Ethanol
(100%)
Cells
Fig. 2. Extraction of 63 Ni 2+ from P. tabaci during cell processing in unfixed (A) and acetic
acid/ethanol-fixed (AA fix) preparations (B). Radioactive levels (derived from scintillation
counts of 1 ml samples) are expressed in relation to 1010 processed cells, and are given for
washing solutions (cross-hatched columns), fixation/dehydration liquids (open columns)
and cell suspensions (solid columns). For each schedule, the cumulative totals (2-11 ^ C i /
10'° bacteria in A; 2-32/iCi/lO 10 bacteria in B) represent the overall levels of radioactivity
present in the samples after the first wash, and are derived by addition of all the separate
supernatant levels plus the cell suspension.
Table 1. Retention of63Ni2+ by cells of P. tabaci after different fixation procedures
Treatment
^iCi/ml*
Untreated
Acetic acid/ethanol
Paraformaldehyde
Glutaraldehyde
3-9XKT 3
1-2X10"3
0-9XKT 3
0-9XKT 3
Mass of Ni/
No. of Ni atoms/
Total bacteria/ml bacterium (nmol)f
bacterium};
4-7X107
4-6X107
4-3X107
4-5X107
12-3X10-"
4-OXlfr"
3-0XKT"
3-Oxlfr"
7-4X104
2-4X104
1-8X104
1-8X104
* Values (derived from scintillation counts of 1 ml samples) are shown for suspensions of cells in
buffer (unfixed) and 100% ethanol (fixed).
f The mass of Ni z+ per bacterium is calculated from the known specific activity of the radioisotope.
J T h e number of atoms is derived using Avogadro's constant.
cell at 16 h. The lower value in the 16 h sample may result from a reduced availability
of 63 Ni 2+ in the incubation medium, or from a reduction in cell uptake as the culture
passes from log to stationary phase.
Wash
3
Wash
2
Fix
(25O/o)
Ethanol
(50%)
Ethanol
(75%)
Ethanol
Ethanol
( 1 00%)
Ethanol
(9OYo)
Cells
Fig. 3. Extraction of 63Ni2+from P. tabaci during cell processing (aldehyde preparations). See the legend to Fig. 2 for details. A . Paraformaldehyde (cumulative total 2.15 &i/lOIO bacteria); B, glutaraidehyde (cumulative total 1 . 9 2 p ~ i / l ~ bacteria).
'0
I
.->
.->
P
iY
a
h
63
Ni2+ uptake in Pseudomonas tabaci
93
Table 2. Retention of63Ni2+ by cells of P. tabaci after different labelling periods
Treatment
time (h)
/iCi/ml*
Total bacteria/ml
Mass of Ni/
bacterium (nmol)
No. of Ni atoms/
bacterium
1
2
8
16
0-4X10"3
1-3X10"3
8-9X10"3
12-6X10"3
S-3X107
12-0X107
86-0X107
240-OX107
1-lxlO" 11
1-6X10"11
l-SxlO""
0-8X10"11
6-7X103
9-7X103
91X10 3
4-8X103
• Values for /id/ml and derived parameters are calculated as for Table 1, for cells labelled over
1-16 h periods. All samples were fixed for 2h in glutaraldehyde, and scintillation counting was
carried out on cell suspensions in 100% ethanol.
Autoradiography
Light- and electron-microscope autoradiography was carried out on 2-h labelled
bacteria fixed separately in acetic acid/ethanol, paraformaldehyde, glutaraldehyde
and glutaraldehyde with osmium tetroxide postfixation.
Light microscopy. Phase-contrast microscopy of 2/im thick sections revealed bacterial aggregations of varying size and packing (Fig. 4A) with each fixation. The
autoradiographic preparations invariably had numerous silver grains over the separate
masses of bacteria, with an appreciable scatter from these aggregates over the
surrounding background (Fig. 4B). Although problems in making reliable counts of
bacteria per unit area made autoradiographic quantitation impossible with these
preparations, simple visual inspection indicated a considerable higher frequency of
silver grains over acetic acid/ethanol preparations (Fig. 4B) compared with those
fixed in either paraformaldehyde (Fig. 5) or glutaraldehyde.
Electron microscopy. Low-power views of bacterial masses in ultrathin section
(Figs 6, 7) confirmed the irregular close-packing seen under the light microscope. At
this magnification, cells fixed in acetic acid/ethanol appeared to have rather featureless, diffuse contents (Fig. 6), while aldehyde-fixed cells (Fig. 7) had a clear central
nucleoid area containing coarse strands and aggregations of electron-dense chromatin.
Autoradiographs processed with D19 developer showed a clear labelling
throughout the bacterial populations, with silver grains (coiled filaments) localized to
Fig. 4. Light-microscope preparations, acetic acid/ethanol-fixed cells, A. Nonautoradiograph. Phase-contrast view of bacterial group, B. Autoradiograph; adjacent section to A. Bright-field view showing dense accumulation of silver grains over bacterial
group, and scatter over adjacent resin area. Bar, 10/im.
Fig. 5. Light microscope autoradiograph (phase-contrast) of paraformaldehyde-fixed
bacteria. The bacterial group is well-labelled, but less heavily than 4B. Bar, 10 (im.
Figs 6—7. Electron-microscope autoradiographs. D19 development. Silver grains are
present as coiled filaments. Both preparations are stained with lead citrate.
Fig. 6. Acetic acid/ethanol preparation. Bar, 0-5/im.
Fig. 7. Glutaraldehyde (3 h) fixation. Strands and accumulations of condensed DNA
are prominent within the central nucleoid (n). Bar, 0-5
R. H. Al-Rabaee and D. C. Sigee
Figs 4-5. For legend see p. 93.
63\r;2
Ni +uptake
95
in Pseudomonas tabaci
\
I-
s
*>
if¥fj
s
--•^Kf-
^^
i*"
>i
%
n
m
•JM*
Figs 6-7. For legend see p. 93.
96
R. H. Al-Rabaee and D. C. Sigee
Table 3. Mean grain counts over electron microscope autoradiographs
Incubation period
(months)
Acetic acid/
ethanol
Paraformaldehyde
Glutaraldehyde
Glutaraldehyde/
osmium tetroxide
2
6
7- 6±2- 5
22- 6±6- 9
2-2 ±0-7
10-2 ±2-8
0
3-Z ± 1-1
0
1-7 ±0-5
Mean grain counts (per 10 bacterial profiles) are given for D19-developed autoradiographs,
processed in two batches at 2 and 6 months after coating with emulsion. Each mean count is derived
from a sample of at least 200 profiles. The grain counts over glutaraldehyde and glutaraldehyde/
osmium tetroxide-fixed cells after 2 months incubation were so low that they did not exceed background. Confidence limits are at 95% level.
individual bacterial profiles (Figs 6, 7). Background labelling (over pure resin) was
normally low, amounting to no more than 1-2 grains per SOOjUm2. Analysis of
autoradiographs processed after exposure periods ranging from 2-6 months showed
a consistent relationship between extent of labelling and type of fixation. Table 3
shows the mean grain count per 10 bacteria in autoradiographs processed at different
periods. Preparations developed after a short (2-month) interval typically had a high
grain frequency over cells fixed in acetic acid/ethanol, fewer grains over
paraformaldehyde-fixed preparations and no appreciable label over glutaraldehydefixed cells. Significant levels of silver grains over the latter were seen only after long
(6-month) periods, when they were considerably less than in the paraformaldehyde
and acetic acid/ethanol preparations. Postfixation in osmium tetroxide (Table 3)
appears to reduce the level of retained 63 Ni 2+ even further. The level of labelling can
also be considered in terms of the percentages of bacteria with 0, 1, 2 etc. associated
silver grains. The histograms shown in Fig. 8 for a 6-month batch of grids show clear
differences between methods of fixation, with 80% labelled profiles in the case of
acetic acid/ethanol fixation, compared to 56% for paraformaldehyde, 32% for
glutaraldehyde and 11 % for glutaraldehyde/osmium tetroxide.
Electron-microscope autoradiographs processed by gold latensification and physical developer (which gives increased emulsion sensitivity; Langford, 1974; Kopriwa,
1975) had more silver grains per bacterial cell with smaller, often clustered silver
grains (Figs 9—11). The silver grains in these preparations were identified as such by
their regular shape and complete electron opacity. They were observed in both stained
(Figs 9, 11) and unstained (Fig. 10) preparations, and some of the larger grains gave
a clear characteristic silver peak when checked by X-ray microanalysis. Although the
occurrence of silver grains outside bacteria was very infrequent (low scatter and
background), non-localized grains over pure resin were occasionally observed. As
with the D19 preparations, highest grain frequencies occurred over the acetic
acid/ethanol preparations (Fig. 9). The uptake of 63Ni2+ did not vary with the
division state of the bacteria, since heavy labelling occurred with both dividing and
non-dividing cells (Fig. 9). In all of the preparations observed, the silver grains
appeared to be largely restricted to the centre of the cells. This was tested statistically
63
Ni2+ uptake in Pseudomonas tabaci
97
Acetic acid/ethanol
302010-
0
1
2
3
5
4
6
40-
7
8
9
10
Paraformaldehyde
302010-
—
^
3
4
1
g
IS
3
70-
0
1
2
5
6
7
60a.
oa.
8
9
10
Glutaraldehyde
5040c
CD
3
CT
CD
302010-
ro
1
CO
i
1
2
i
3
i
4
i
5
i
6
t
7
i
8
i
9
(
10
8070-
Glutaraldehyde/osmium tetroxide
6050403020101
2
3
4
5
6
7
8
9
Number of silver grains per bacterium
10
Fig. 8. Bacterial labelling (EM autoradiographs). The frequency of bacterial profiles with
0, 1, 2 or more silver grains are shown for D19-processed autoradiographs of cells fixed in
four ways. All the preparations were processed at the same time, so had identical conditions of development and photographic fixation. Each histogram is derived from a
sample of at least 200 profiles.
R. H. Al-Rabaee and D. C. Sigee
%h.
—t
.
•
-
l
* «*
Figs 9-10
<
63
Ni2+ uptake in Pseudomonas tabaci
99
in the acetic acid/ethanol-fixed preparations by comparing the distances between
individual silver grains and the nearest point on the bacterial boundary with
equivalent measurements in an equal sample of random points (Fig. 12). The two
histograms seen in Fig. 12 appear to be quite distinct, and comparison of the means
on a null hypothesis confirms that the distributions of silver grains and random points
are significantly different.
High-power examination of acetic acid/ethanol-preserved cells reveals a surprising
amount of detail for a fixative that is generally regarded as unsuitable for electron
microscopy. The central nucleoid appears as a system of interconnecting spaces
ramifying throughout the ribosomal groundplasm, and contains diffuse chromatin,
which is condensed at various sites in the bacterium as electron-dense patches (Fig.
11). A high proportion of the silver grains shows clear localization to these regions of
condensed chromatin (Fig. 11). Silver grains were also observed over apparently clear
areas of nucleoid and over ribosomal groundplasm, but were hardly ever seen over the
bacterial cell wall.
In aldehyde-fixed cells, with a more clearly defined central nucleoid space, the fine
silver grains showed an even clearer localization to the chromatin-containing region
of the cell (Fig. 10). Very few grains occurred over surrounding ribosomal groundplasm or cell wall.
DISCUSSION
Addition of 63Ni2+ to the bacterial cultures used in these experiments resulted in
an overall nickel concentration of 1 -2x 10~5 M (no detectable Ni in the original growth
medium). This level of Ni had no limiting effect on the multiplication of bacteria
(population increase similar to control without nickel), so that the uptake of 63 Ni 2+
in this investigation was occurring in non-toxic conditions.
The scintillation counts of the processing solutions indicate that most of the incorporated 63 Ni 2+ is water-soluble. This soluble Ni is present partly as a readily diffusible
form (with approximately 50 % being lost during washing) and partly as an extractable form (with a further 30% being lost during fixation and dehydration). These
Figs 9—11. Electron-microscope autoradiographs. Gold latensification and physical
development (1-2min). Silver grains (Ag) are present as small electron-dense granules.
Fig. 9. Acetic acid/ethanol preparation. Stained with alkaline lead citrate after
photographic processing. The mean grain size in this preparation is 0-04 /^m. Silver grains
are present over both dividing (d) and non-dividing bacteria, and over all angles of section.
t, transverse section; /, longitudinal section. Bar, 0-5 fan.
Fig. 10. Glutaraldehyde preparation. Unstained. These cells were labelled with M Ni
for 8h, and fixed for 2h. n, central electron transparent nucleoid containing chromatin
reticulum; r, peripheral ribosomal groundplasm. The silver grains (Ag) have a mean
diameter of 0-03 fun. Bar, 0-2 fim.
Fig. 11. Acetic acid/ethanol preparation. Stained. High-power view showing bacteria
with regions of condensed chromatin (arrows), surrounded by electron-transparent
nucleoid spaces (s) and ribosomal groundplasm (r). Silver grains (Ag) show a clear localization to the condensed chromatin. Bar, 0-2 fim.
m
R. H. Al-Rabaee and D. C. Sigee
Fig. 11. For legend see p. 99.
a
Ni2+ uptake in Pseudomonas tabaci
101
20-
CO
§ 15
a
|
Z
5-
_D •_
_n
i
005
0-1
005
0-2
0-15
0-1
Distance to edge of bacterium
0-15
0-2
0-25
0-3
0-35
0-25
0-3
0-35
15 3
I
o 10C
a
i
Fig. 12. Distribution of silver grains and random points over bacteria, A. Autoradiograph
(EM). Acetic acid/ethanol-fixation. Physical developer. The position of 221 silver
grains over 40 bacteria was measured individually as the distance to the nearest point on
the cell boundary, and expressed cumulatively as a histogram. x = 0-213; i = 0-05;
n = 221;
z =
=16-3
B. Random point distribution; 221 random points were plotted over tracings of the bacterial profiles used in A, and a corresponding histogram was plotted. /io = 0'132;
s' = 0-074; n' = 221.
The two histograms appear quite distinct. Comparison of the means (x,fio) on a null
hypothesis confirms this. With standard deviations (s, s') and sample size (n) as shown,
the resulting value forz of 16-3 exceeds the theoretical value of zoo\ = 2-33.
102
R. H. Al-Rabaee and D. C. Sigee
experiments do not give any indication of the location of this soluble nickel in the cell,
though adsorption of cations at the cell'surface (Corpe, 1975) or cell wall (Beveridge
& Murray, 1976; Haavik, 1976), as well as uptake into the protoplast cytosol, are
possibilities. The rapid and continued labelling of bacterial cells during the log growth
phase is consistent both with a continuous adsorption onto newly formed cell-wall
material, and with an active cation-transport system into the bacterial protoplast,
similar to that observed in Bacillus megaterium (Schneider, 1977) and Escherichia
colt (Jasper & Silver, 1977).
The Ni2+ that remains in fixed, dehydrated cells represents the bound, insoluble
nickel, and amounts to roughly 10—20% of the total nickel initially present. The use
of coagulative (acetic acid/ethanol) and additive (aldehyde) fixatives in determining
insoluble Ni content was important in terms of both possible fixation artefacts and the
exact level retained. The possibility of artifactual binding of radioactive molecules has
been shown, for example, by Peters & Ashley (1967), where radioactive amino acids
were bound to cell proteins by glutaraldehyde but not paraformaldehyde fixation. The
retention of 63 Ni 2+ in bacteria fixed by all three types of fixative suggests that the
bound (insoluble) nickel is not simply present as a fixation artefact.
Both the scintillation and autoradiographic studies show that higher levels of
63
Ni2+ were retained with acetic acid/ethanol fixation than with aldehyde fixation.
Previous work with other cells (Sigee & Kearns, 1982) has suggested that coagulative
fixation preserves more medium MT proteins than does aldehyde fixation, and the
additional "Ni2"1" seen in bacterial cells after acetic acid/ethanol treatment may reflect
its association with these proteins.
The comparison between glutaraldehyde and paraformaldehyde fixations
represents a major difference between the scintillation and autoradiographic results.
In the scintillation counts, these two fixatives led to equal retention of the 63 Ni 2+ ,
while in the autoradiographs the level of label in the paraformaldehyde preparations
was considerably greater than with glutaraldehyde. There is no immediate explanation for these results, though the difference may be due to differences in fixation time,
which was 2h in the scintillation studies and 3 h in the autoradiography. The extra
fixation time does lead to a greater condensation of the bacterial chromatin (compare
Figs 7 and 10), and there is a possible extra, and differential, loss of 63 Ni 2+ with the
two fixations during the extra hour.
Both scintillation counting and autoradiography are highly sensitive methods for
the detection of 63 Ni 2+ . It is interesting to note in this context that X-ray microanalysis
was not sufficiently sensitive to detect incorporated Ni in any of the sections. Scintillation methods have been used to determine the uptake of Ni in a variety of organisms,
with a substantial amount of work being carried out on bacteria (Tabillion & Kaltwasser, 1977; Diekert, Weber & Thauer, 1980; Friedrich, Schneider & Friedrich,
1982; Jarrell & Sprott, 1982). Autoradiography, on the other hand, has been much
less frequently used. Previous studies include light-microscope work by Oskarsson &
Tjalve (1979, 1980) on ^NiCh uptake in mice, and light- and electron-microscope
studies by Sigee (1982) on 63Ni2+ uptake in dinoflagellates, but there appears to be no
previously published work involving bacteria. This is surprising in view of the
63
Ni2+ uptake in Pseudomonas tabaci
103
63
2+
apparent suitability of this isotope for autoradiography, since Ni has a long halflife (92 years), the radiodecay results entirely in soft beta emission (Kirby, 1961),
and the label can be obtained at high specific activity. The higher mean energy of
emission compared to tritium (67keV compared to 18keV) implies a lower
resolution than normally obtained with this technique, though this was only apparent at the level of the light microscope, where a wide scatter of silver grains occurred
around heavily labelled bacterial groups. At the level of the electron microscope,
the use of gold latensification followed by physical development (Salpeter & Bachmann, 1964; Kopriwa, 1975) provides a very powerful technique for determining
localization of label within intact cells, resulting in high emulsion sensitivity, small
silver grains (high resolution) and low background. The multiple silver grains
obtained with this technique probably arise by partial development of several latent
images within a single silver halide crystal (Mees, 1954; Langford, 1974), arising
from a single beta particle sensitization. This is in contrast to chemical (D19)
development, where normally not more than one silver grain arises from a single
sensitization. For this reason, grain counts were made from the D19-processed
autoradiographs, while studies on localization were made in the physically developed
preparations.
In both the acetic acid/ethanol- and aldehyde-fixed cells, there was a high degree
of localization of bound 63Ni2+ to the central nucleoid region of the cell. Silver grains
were conspicuously absent from the cell surface, and the cell wall, and were infrequently seen over peripheral ribosomal groundplasm (aldehyde preparations). In
the centre of the cell the extent of the nucleoid space and the appearance of the
chromatin varied considerably with fixation. This variation proved quite useful, since
the discrete central nucleoid in aldehyde-fixed cells permitted a defined localization
of 63 Ni 2+ uptake to this part of the cell, while the precipitated masses of chromatin in
the acetic acid/ethanol preparations allowed clear localization of label to this region
within the nucleoid.
It seems likely (Haggis & Bond, 1981) that the nucleoid regions seen in ultrathin
sections of fixed cells (Ryter & Kellenberger, 1958) and in whole living cells (Binnerts, Woldringh & Brakenhoff, 1982) contain only part of the bacterial DNA. Ryter
& Chang (1975) have suggested from autoradiographic studies that the nucleoid D NA
contains only inactive genes, while active genes involved in transcription extend into
the ribosomal groundplasm. Studies of isolatedE. coli nucleoids (Kavenoff & Bowen,
1976; Pettijohn, 1976) confirm the presence of a core region with lateral loops spreading out, possibly corresponding to central and peripheral (ribosomal) regions in the
intact cell. In view of these observations, the studies on 63 Ni 2+ uptake presented here
suggest that incorporation of label occurs mainly into the core (genetically inactive)
DNA, and that the genetically active DNA fibrils in the peripheral part of the cell have
little associated radioactivity. The association of divalent cations with nucleic acids
has been widely documented (for discussion, see Kearns & Sigee, 1980; Sigee,
19836). It is proposed that, among other functions, these cations are important in
stabilizing the in vivo structure of the DNA, particularly in the absence of associated
histones (cationic non-histone stabilization; Sigee, 1983a,6).
104
R. H. Al-Rabaee and D. C. Sigee
The distinction between genetically active and inactive DNA seen in bacteria shows
a remarkable resemblance to the situation in dinoflagellates, the only eucaryotic group
with a complete lack of chromatin histones (Rizzo & Nooden, 1974). In
dinoflagellates, as in the bacteria studied here, divalent cations are associated particularly with the genetically inactive DNA (in permanently condensed
chromosomes) (Kearns & Sigee, 1979, 1980; Sigee & Kearns, 1980, 1981), while the
genetically active extrachromosomal filaments appear to have much lower levels
(Sigee, 1984). The results obtained in this investigation thus lend some support to the
suggestion (Sigee, 1983a,fc) that cationic non-histone stabilization of dinoflagellate
DNA resembles the situation in bacteria, and might therefore be regarded as
primitive.
REFERENCES
BEVERIDGE, T. J. & MURRAY, R. G. (1976). Uptake and retention of metals by cell walls of Bacillus '
subtilis.J.Bact. 127, 1502-1518.
BINNERTS, J. S., WOLDRINGH, C. L. & BRAKENHOFF, G. J. (1982). Visualization of the nucleoid
in living bacteria on polylysine-coated surfaces by the immersion technique. J. Microsc. 125,
359-363.
BROWN, S. S. & SUNDERMAN, F. W. (1980). Nickel Toxicology. London: Academic Press.
CORPE, W. A. (1975). Metal-binding properties of surface materials from marine bacteria. Dev.
Ind. Microbiol. 16, 249-255.
DlEKERT, G., WEBER, B. & THAUER, R. K. (1980). Nickel dependence of factor F43o content in
Methanobacteriiim thermoautotrophicum. Arch. Mikrotnol. 127, 273—278.
DULKA, J. J. & RISBY, T. H. (1976). Ultratrace metals in some environmental and biological
systems. Analyt. Chem. 48, 640-653.
FRIEDRICH, C. G., SCHNEIDER, K. & FRIEDRICH, B. (1982). Nickel in the catalytically active
hydrogenase of Alcaligenes eutrophus.J. Bact. 152, 42-48.
HAAVIK, H. I. (1976). On the role of bacitracin peptides in trace metal transport by Bacillus
licheniformis.J. gen. Microbiol. 96, 393-399.
HAGGIS, G. H. & BOND, E. F. (1981). A new approach to the study of the E. coli nucleoid. J .
Microsc. 122, 15-22.
HUTCHINSON, T. C. (1981). Effects of Nickel in the Canadian Environment. Ottawa: NRCC
publication 18568.
JARRELL, K. F. & SPROTT, G. D. (1982). Nickel transport in Methanobacteriiim bryantii.J. Bact.
151, 1195-1203.
JASPER, P. & SILVER, S. (1977). Magnesium transport in microorganisms. In Microorganisms and
Minerals (ed. E. D. Weinberg), pp. 7-47. New York: Dekker.
JAWORSKI, J. F. (1974). Relative sensitivity calibration in spark-source mass spectrometry. Ph.D.
thesis, Cornell University, Department of Chemistry.
KASPRZAK, K. S. & SUNDERMAN, F. W. (1979). Radioactive w Ni in biological research. PureAppl.
Chem. 51, 1375-1389.
KAVENOFF, R. & BOWEN, E. F. (1976). Electron microscopy of membrane-free folded
chromosomes from E. coli. Chromosoma 59, 89-101.
KEARNS, L. P. & SIGEE, D. C. (1979). High levels of transition metals in dinoflagellate
chromosomes. Experientia 35, 1332—1333.
KEARNS, L. P. & SIGEE, D. C. (1980). The occurrence of Period IV elements in dinoflagellate
chromatin: an X-ray microanalytical study. J. Cell Sci. 46, 113-127.
KIRBY, L. J. (1961). The Radiochemistry of Nickel. U.S.A.: National Academy of Sciences.
KOPRIWA, B. M. (1975). A comparison of various procedures for fine grain development in electron
microscope autoradiography. Histochemistry 44, 201-224.
LANGFORD, M. J. (197'4). Advanced Photography. London: The Focal Press.
MEES, C. E. (1954). The Theory of the Photographic Process. Toronto: Macmillan Press.
63
Ni2+ uptake in Pseudomonas tabaci
105
w
OSKARSSON, A. & TJALVE, H. (1979). An autoradiographic study on the distribution of NiCl2 in
mice. Ann. din. Lab. Sci. 9, 47-51.
OSKARSSON, A. & TJALVE, H. (1980). Effects of diethyldithiocarbamate on the tissue distribution
of H NiCl 2 in mice. Archs Toxicol. 45, 45-52.
PETERS, T. & ASHLEY, C. A. (1967). An artefact in radioautography due to binding of free amino
acids to tissues by fixatives. J. Cell Biol. 33, 53-60.
PETTIJOHN, D. E. (1976). Prokaryotic DNA in nucleoid structure. Crit. Rev. Biochem. 4, 175-202.
Rizzo, P. J. & NOODEN, L. D. (1974). Partial characterisation of dinoflagellate chromosomal
proteins. Biochim. biophys. Ada 349, 415-427.
RYTER, A. & CHANG, A. (1975). Localization of transcribing genes in the bacterial cell by means
of high-resolution autoradiography. J. molec. Biol. 98, 797-810.
RYTER, A. & KELLENBERCER, E. (1958). Etude en microscope electronique de plasmas contenant
de l'acide desoxyribonucleique. Z. Naturf. 13, 597-605.
SALPETER, M. M. & BACHMANN, L. (1964). Autoradiography with the electron microscope. J'. Cell
Biol. 22, 469-477.
SCHNEIDER, J. (1977). Speicherung und Hemmstoffwirkung von Nickel in aeroben Bakterien.
Diplomarbeit, Universitat Saarbrucken.
SIGEE, D. C. (1982). Localised uptake of "Nickel into dinoflagellate chromosomes: An
autoradiographic study. Protoplasma 110, 112-120.
SIGEE, D. C. (1983a). Some observations on the structure, cation content and possible evolutionary
status of dinoflagellate chromosomes. Bot.J. Linn. Soc. (in press).
SIGEE, D. C. (19836). Detection and quantitation of divalent cations in dinoflagellate chromatin.
SEM 1983 IV, 1779-1793.
SIGEE, D. C. (1984). Genetically active DNA and inactive DNA in dinoflagellate chromatin.
Biosystems 16, 203-210.
SIGEE, D. C. & KEARNS, L. P. (1980). Detection of nickel in the chromatin of dinoflagellates by
X-ray microanalysis. In Nickel Toxicology (ed. S. S. Brown & F. W. Sunderman), pp. 00—00.
London: Academic Press.
SIGEE, D. C. & KEARNS, L. P. (1981). Nuclease extraction of chromosome-bound metals in the
dinoflagellate Glenodinium foliaceum: an X-ray microanalytical study. Cytobios 31, 49—65.
SIGEE, D. C. & KEARNS, L. P. (1982). Differential retention of proteins and bound divalent cations
in dinoflagellate chromatin fixed under varied conditions: An X-ray microanalytical study.
Cytobios 33, 51-64.
SUNDERMAN, F. W. (1978). A commentary on the 1978 Kristiansand Conference on nickel
Toxicology. Ann. din. Lab. Sci. 8, 491—494.
SUNDERMAN, F. W. (1980a). Analytical biochemistry of nickel, Pure Appl. Chem. 52, 527-544.
SUNDERMAN, F. W. (19806). Nickel. In Methods ofAnalysis for Heavy Metals and Human Toxicology (ed. A. Vercruysse). Amsterdam: Elsevier Scientific Publishing Co.
TABILLION, R. & KALTWASSER, H. (1977). Energieabhangige MNi-Aufnahme bei Alcaligenes
eutrophus Stamm HI und H16. Arch. Mikrobiol. 113, 145-151.
WILLIAMS, M. A. (1977). Autoradiography and immunocytochemistry. In Practical Methods
in Electron Microscopy (ed. A. M. Glauert), vol. 6, pp. 77-155. Amsterdam: North-Holland
Publishing Co.
(Received 19 October 1983-Accepted 10 February 1984)