Biol Fertil Soils (2008) 44:557–569
DOI 10.1007/s00374-007-0235-5
ORIGINAL PAPER
Diversity of heterotrophic aerobic cultivable microbial
communities of soils treated with fumigants and dynamics
of metabolic, microbial, and mineralization quotients
Stefano Mocali & Donatella Paffetti &
Giovanni Emiliani & Anna Benedetti & Renato Fani
Received: 20 June 2006 / Revised: 31 July 2007 / Accepted: 1 August 2007 / Published online: 6 September 2007
# Springer-Verlag 2007
Abstract A combination of molecular and classical techniques was used to study the composition, structure, diversity, and dynamics of an aerobic heterotrophic cultivable
bacterial community isolated from five different soil samples
treated with the fumigant agent 1,3-dichloropropene (1,3-D)
and further subjected to nitrogen–phosphorous–potassium
(NPK) fertigation (F), amendment (C2 and C4), and NPK
fertigation plus amendment (F+C) in two different periods
(May and July). The restriction and sequence analysis of 16S
rDNA from 189 isolates revealed a very high percentage
(94%) of Gram-positive bacterial isolates, most of which
(83%) belonging to the genus Bacillus. The degree of
intraspecific genetic diversity was high, as shown by random
amplified polymorphic DNA analysis. These data seem to be
related with the increase in microbial biomass C (Cmic)
content and the decrease in the total organic C (Corg) and
metabolic quotient (qCO2) values, especially in amended
Stefano Mocali and Donatella Paffetti contributed equally to this
work.
S. Mocali : A. Benedetti
C.R.A.-Istituto Sperimentale per la Nutrizione delle Piante,
Via della Navicella 2-4,
00184 Rome, Italy
D. Paffetti : G. Emiliani
Dipartimento di Scienze e Tecnologie Ambientali Forestali,
University of Florence,
Via S. Bonaventura, 13,
50145 Florence, Italy
R. Fani (*)
Dipartimento di Biologia Animale e Genetica,
University of Florence,
Via Romana 17-19,
50125 Florence, Italy
e-mail: r_fani@dbag.unifi.it
soils (C2, C4) where soil microflora mineralized the organic
matter of the added fertilizers. In a short term, it is suggested
that the presence of very high percentage of Gram-positive
bacteria might be related to the ability of these bacteria to
form spores so as to be resistant to fumigants rather than
being the result of a selective pressure in the predominance
of microbial species with a set of genes involved in
biodegradation of 1,3-D.
Keywords Biodiversity . Fumigants . Soil fertility
Introduction
One of the most important problems in crop production is
represented by parasites of plants such as nematodes
(Baldwin et al. 2004). In particular, the nematode Heterodera
carotae causes damages in carrots and in some other
Umbrelliferae. For many years, the fumigant methyl
bromide (MeBr) has been used to provide an effective
control of nematodes in agricultural soils. However, it was
recognized to have negative effects on stratospheric ozone
(Park et al. 2004; Prather et al. 1984); hence, according to
the protocol of Montreal (1997), it was phased-out of use in
Europe and in USA on January 1, 2005. Most of the
fumigants commonly used to replace MeBr can have broad
biocidal activity in agricultural soils (Giannakou et al. 2002),
but their effects on soil microflora are still largely unknown.
Repeated application of fumigants to agricultural soils for
many years may modify soil microbial diversity, biomass,
and activities with indirect effects on soil “quality” that,
according to Doran and Parkin (1994), can be defined as
“the capacity of the soil to interact with the ecosystem in
order to maintain biological productivity, environmental
quality and to promote animal and plant health.”
558
Recently, the effects of 1,3-dichloropropene (1,3-D),
methyl isothiocyanate, and chloropicrin on the structure of
soil microbial community have been studied using both
culture-dependent and culture-independent approaches
(Ibekwe et al. 2001a). It was shown that Gram-positive
bacteria survived better than Gram-negative to the fumigant
treatment and that 1,3-D was the least effective fumigant in
affecting the structure of soil microbial community. Ibekwe
et al. (2001b) also showed that, after treating microcosm
with 1,3-D for 6 months, Pseudomonas and Actinomadura
were the predominant soil bacterial species. Furthermore,
Smelt et al. (1989) and Ou et al. (1995) found that, after
repeated applications of 1,3-D, the effect of the fumigantnematocide was markedly decreased due to the microbial
degradation that probably was also stimulated by the application of compost manure (Gan et al. 1998). On the other
hand, very little information on the effect of the addition of
different amendments on microbial activity and soil quality
of fumigated soils is available.
Because biological parameters have assumed particular
importance in the assessment of soil quality, because
organisms respond more rapidly than most chemical and
physical parameters to changes in land use, environmental
conditions, or contamination (Burns et al. 2006), a number
of soil criteria related with soil microbiological functions,
such as the metabolic quotient (qCO2), the microbial
quotient (Cmic/Corg ×100), and the mineralization quotient
(qM), have been used to evaluate the efficiency of
utilization and conservation of organic matter after different
agronomical treatments. Indices based on the combination
of microbial respiration and microbial biomass measurement, such as qCO2, may be more sensitive than microbial
activity or microbial biomass alone to assess effects of soil
pollution (Nannipieri et al. 1990; Brookes 1995; Dilly and
Munch 1998). The qCO2 can reflect the maintenance of
carbon requirement of cells, and an increase of qCO2 can be
related to microbial stress (Anderson 2003) or to changes in
the bacterial/fungal ratio (Sakamoto and Oba 1994; Landi
et al. 2000). The microbial quotient is a sensitive parameter
of changes in the organic C content (Anderson and Domsch
1990) and usually ranges between 1 and 4% with values
below 1% being indicative of microbial stress (Jenkinson
and Ladd 1981). The mineralization quotient representing
the ratio between respiration and the total organic C content
indicates the efficiency of microflora in metabolizing
organic matter.
To date, no study concerning with the effect of nematode
fumigant on the abovementioned ratios is available, and
this work may help in understanding better soil functionality after the fumigant treatment, by integrating data
Biol Fertil Soils (2008) 44:557–569
regarding “soil quality” with taxonomical analysis of
cultivable bacteria.
The main objective of this work was to evaluate the
effect of the 1,3-D fumigant on microbial activity, microbial
biomass, and diversity of heterotrophic culturable microbial
communities of soil, although they represent a minor
proportion of bacteria inhabiting soil. We wanted also to
evaluate the effects of different agronomical treatments on
soil recovery in cases of negative effects by the fumigant.
Materials and methods
Soil treatments and sample collection
This study was conducted in the agricultural cooperative S.
Antonio of Maccarese, near Rome (Italy), one of the most
important agricultural areas for carrots (Daucus carota).
The soil had a loamy-sandy texture (IUSS) with pH 8.3 and
a high content of exchangeable Ca (2.74 cmol(+) kg−1). The
content of total N and organic matter was very low (0.04
and 0.43%, respectively) as well as the cationic exchange
capacity (7.11 meq/100 g). This soil has been fumigated for
more than 20 years with 1,3-D against nematodes. Samples
were collected from the top 20 cm of soil in May 2002
(40 days after sowing) and at the end of July 2002 (after
harvesting) with a spade because the soil was too sandy
(92%) for a hole digger. The field experiment was based on
the following four treatments, which were carried out
adopting a randomized-block experimental design in five
plots with three replicates, with a total of 15 different plots:
1. F; fertigation with 100 kg of ammonium nitrate applied
before sowing, 50 kg of ammonium nitrate added at the
end of May, and 50 kg of potassium nitrate added at the
end of June.
2. F+C; soil was amended with 4 t of mixed compost
amendment ha−1 before sowing and fertigated at the
end of May with 100 kg of ammonium nitrate.
3. C2; soil was amended with 16 t of mixed compost
amendment ha−1 before sowing (in February).
4. C4; soil was amended with 32 t of mixed compost
amendment ha−1 before sowing (in February).
5. T is the control (not treated) soil.
The compost (C; N content of 25%) was produced
aerobically by composting a mixture of residual borlande
from the distillation of grape cake (50–60%), olive cake
(10–15%), residual of pruning of olive, grape vine, and
other fructiferous (25–40%).
Biol Fertil Soils (2008) 44:557–569
Each treatment was applied in plots of 5×5 m located at
a distance of 1.5 m from each other, and for each of the 15
plots, 4 subsamples were collected at each sampling period
(May and July), air-dried, and sieved (2 mm). Subsamples
of the same plot were then pooled, and the resulting sample
was mixed and stored for subsequent analyses.
Biochemical and microbial parameters
The soil used for this experiment was analyzed according to
the official methods of Ministero delle Politiche Agricole e
Forestali (1997, 1999). The total organic carbon (Corg),
expressed as mg kg−1, was determined according to
Springer and Klee (1954). Soil respiration was expressed
as mg C-CO2 Kg−1soil d−1 and measured in a closed
environment after 1, 2, 4, 7, 10, 14, 17, and 21 days
according to Isermeyer (1952). Soil respiration data were
fitted according to the first-order exponential equations Ct =
C0(1−e−kt), representing the cumulative mineralization
curves, where Ct is the total amount of mineralized carbon
in laboratory conditions during 21 days of analysis (Riffaldi
et al. 1996). A nonlinear regression square analysis was
used to calculate the potentially mineralizable carbon pool
C0 [mg(C) kg−1 soil] for each investigated soil. Basal
respiration (Cbas) represents the value of mineralized C in a
definite period of time (21 days), when steady-state
condition has been reached. Microbial biomass carbon Cmic
(mg C kg−1 soil) was measured by the fumigation–
extraction method (Vance et al. 1987) on air-dried soils
which were conditioned at −33 kPa water tension and
preincubated for 10 days in open glass jars at 30°C. Several
derived parameters were then calculated: the qCO2 (metabolic quotient), calculated by the ratio Cbas/Cmic [(mg CCO2 basal mg Cmic−1) h−1], the ratio Cmic/Corg (microbial
quotient) expressed as a percentage [mg C biomass/mg total
organic C × 100], and qM (mineralization quotient),
expressed as mg C-CO2 kg−1 soil h−1 and calculated from
cumulative respiration after 21 days by the relationship: CCO2 cum/Corg.
Bacterial counts
Bacteria were extracted from fresh soil and diluted in saline
solution (0.85% NaCl) at suitable concentrations (10−1,
10−2, 10−3, 10−4, 10−5, 10−6, 10−7). Aliquots (0.1 ml) of
each dilution were plated in triplicate onto Luria–Bertani
(LB) medium. Plates were incubated at 28°C for 2 days;
thereafter, colonies were counted, and the results reported
in Table 2 expressed as n° of bacteria/g of fresh soil.
Twenty visually distinct colonies were picked up at random
559
from each sample at each sampling date and transferred to
Multiwell plates (Corning, New York, NY, USA). Each
colony was purified on Nutrient Starch Agar (NSA) and
stored at −20°C in LB (Difco Laboratories) with 20%
glycerol.
PCR amplification, restriction, sequencing, and analysis
of bacterial 16S rDNA
Bacterial colonies grown overnight on LB plates were
resuspended in 20 μl sterile distilled water, heated to 95°C
for 10 min, and cooled on ice for 2 min; each lysate (2 μl)
was used for the amplification reaction via polymerase
chain reaction (PCR). Amplification of 16S rRNA gene
was performed in a total volume of 20 μl containing 2 μl of
10× reaction buffer (Polymed, Italy), 1.5 mM MgCl2,
150 ng of each primer [27f, 5′ GAGAGTTTGATCCTGG
CTCAG, and 1495r, 5′ CTACGGCTACCTTGTTACGA]
(Grifoni et al. 1995), 200 μM of each dNTP, 0.5 U of Taq
DNA polymerase (Polymed). The reaction mixtures, after
incubation at 95°C for 90 s, were subjected to the following
temperature cycle: denaturation at 94°C for 30 s, annealing
temperature for 30 s, extension at 72°C for 4 min; the
annealing temperature was 60°C for the first five cycles,
55°C for the following five cycles, and 50°C for the last 25
cycles. Then the mixtures were incubated at 72°C for
10 min; 2 μl of each amplification mixture was analyzed by
agarose gel (1.2% w/v) electrophoresis in Tris–acetate–
ethylene diamine tetraacetic acid (TAE) buffer (0.04 M
Tris–acetate, 0.001 M EDTA) containing 0.5 μg/ml (w/v)
ethidium bromide.
For restriction analysis, amplified 16S rRNA gene
(1.5 μg) was digested with 3 U of the endonuclease AluI
(Boehringer, Mannheim, Germany) in 30 μl, for 3 h at 37°C.
The enzyme was then inactivated at 65°C for 10 min.
Reaction products were resolved by agarose gel (2.5% w/v)
electrophoresis in TAE buffer and stained with 0.5 μg/ml
(w/v) ethidium bromide.
The band of interest (observed under UV, 312 nm) was
excised from the gel and purified using the “QIAquick” gel
extraction kit (QiAgen, Chatsworth, CA, USA) according
to manufacturer’s instructions. Direct sequencing was
performed on both DNA strands using an ABI PRISM
310 Genetic Analyzer (Applied Biosystems) and the
chemical dye terminator (Sanger et al. 1977).
Nucleotide sequences were retrieved from the GenBank,
EMBL, and RDP databases. BLAST probing of the DNA
databases was performed with the BLASTN option of the
BLAST program (Altschul et al. 1997). The ClustalW
program (Thompson et al. 1994) was used to align the 16S
560
rRNA sequences obtained with the most similar ones
retrieved from the databases. Each alignment was checked
manually, corrected, and then analyzed using the neighborjoining method (Saitou and Nei 1987) according to the
model of Kimura 2-parameter distances (Kimura 1980).
Phylogenetic trees were constructed with the aligned
sequences using Molecular Evolutionary Genetics Analysis
3 software (Kumar et al. 2004). The robustness of the
inferred trees was evaluated by 500 bootstrap resamplings.
Biol Fertil Soils (2008) 44:557–569
computes the statistical significance of each partition. This
method allows computing the proportion of the overall
community variation, which accounts for the differences between groups formed, joining together isolates from different
samplings. Analyses were only performed on haplotype
frequencies and for each analysis; 16,000 permutations were
computed to obtain the significance levels of the variance.
Results
RAPD fingerprinting
Soil properties
Random amplification of DNA fragments was carried out
in 25 μl of Platinum buffer (Gibco BRL) containing 3 mM
MgCl2, 2 μl of lysed cell suspension, 500 ng of primer
AP12 (5′-CGGCCCCTGC-3′), 200 mM of each dNTP, and
0.625 U of Platinum Taq polymerase (Gibco BRL; Mori
et al. 1999). The reaction mixtures were incubated in a
thermal cycler (model 9600; Applied Biosystems) at 94°C
for 2 min. They were then subjected to 45 cycles, each
consisting of incubation at 95°C for 30 s, 36°C for 1 min,
and 72°C for 2 min; finally, the reactions were incubated at
75°C for 10 min and then at 60°C for 10 min. Reaction
products were analyzed by agarose (2%, w/v) gel electrophoresis in TAE buffer containing 1 μg/ml of ethidium
bromide.
Using fingerprinting pattern of each random amplified
polymorphic DNA (RAPD) product, genetic similarities in
the different samples belonging to the same haplotype
group were determined by pairwise comparison of the
presence and absence of bands with Gelcompare II software
(Applied Maths). A matrix containing similarity values was
obtained with the Dice coefficient. This matrix was used to
construct a dendrogram according to the unweighted-pair
group method, using arithmetic average (UPGMA) cluster
analysis.
Statistical analysis
Soil biochemical and microbiological data were tested for
homogeneity of variance with the Levene test before
undergoing the analysis of variance (ANOVA) using the
program SPSS for Windows (vers. 11.0, 2001). After
obtaining a statistically significant F test from the ANOVA,
a post hoc comparison method (Tukey HSD) was used to
identify which groups are particularly different from each
other. To assess the partitioning of the genetic variance
between different groups of communities, the analysis of
molecular variance (AMOVA; Excoffier et al. 1992) using
the program ARLEQUIN 2.000 (Schneider et al. 1997) was
used. AMOVA divides the total variance into three hierarchical partitions, that is among groups, among populations
within groups, and within populations. A test of permutation
Samples collected in May from plots T and F (Table 1) did
not show significant differences in any chemical or microbiological parameter (p>0.05). As expected, the fertilization
with mixed C2 and C4 compost produced a significant
increase in the Corg content (+46.1% and +75%, respectively; p<0.05) and in the basal respiration (+112.8% and
+250.5%, respectively; p<0.05) with respect to values of T.
Moreover, a consistent increase in Cmic values in F+C
(+169.9%), C2 (+121.6%), and C4 (+207.6%) samples and
in C0 values in C2 (+135.3%), C4 (+210.6%), and F+C
(+88.6%) samples with respect to values of the untreated
control (p<0.005) was observed. The metabolic quotient
was the lowest in the F+C sample, but differences among
soil samples were not significant, whereas the mineralization quotient in C2, C4, and F+C samples was significantly
higher than values of F and T samples. The highest values
of the microbial quotient (Cmic/Corg ×100) were shown by
soils amended with F+C (+134.6%; p<0.05) with respect
to the value of the control plots, whereas C2 and C4 samples
showed values higher than F and T samples but lower than
the value of the F+C sample (p<0.05).
In July (Table 1), the content of the total organic carbon
generally showed lower values than in May, but in C2 and
C4 plots, it was still higher than in F+C soils (+19.6%, p<
0.05 and +34.7%, p<0.001, respectively); the T and F soils
showed the lowest values. In comparison to May,
Corg values decreased more in T (−44%), F (−41%), C2
(−27%), and C4 (−32%) than in F+C soils (−23%). Basal
respiration was not significantly different between different
soils (p>0.05), but C4 samples maintained high values of
C0.
Microbial biomass of C2 and C4 plots was greater than in
any other sample (p<0.001), and it was significantly higher
in July than in May (p<0.05), while a strong decrease in F+
C soil was observed (−33.6%). The metabolic quotient of
C2 and C4 soils was significantly lower than in control soils
(p<0.05), while the qCO2 of the F+C soils showed
intermediary values between F and T soils.
The mineralization quotient of the T soil was significantly higher than the other treated soils (p<0.05). Finally,
Biol Fertil Soils (2008) 44:557–569
561
Table 1 Chemical and microbiological parameters of soils (Corg total organic C, Cbas basal respiration, C0 potentially mineralizable C, Cmic
microbial biomass C, qCO2 metabolic quotient, qM mineralization quotient, and Cmic/Corg contribution of microbial biomass carbon to soil
organic carbon)
Parameter
Units
May
Corg
g C/100 g soil
Cbas
mg(C-CO2)
kg−1soil
(21st day)
mg(C-CO2)
kg−1soil
mg C kg−1soil
C0
Cmic
qCO2
qM
Cmic/Corg
[(mg C-CO2 bas
mg−1 Cmic) h−1]
104
(mg C-CO2
mg−1C)
mg Cmic mg−1 C
July
T
F
F+C
C2
C4
T
F
F+C
C2
C4
0.52a
(0.06)
1.94a
(0.28)
0.60a
(0.04)
2.56a
(0.61)
0.60a
(0.02)
3.93a
(0.99)
0.76b
(0.03)
4.13ab
(0.92)
0.91c
(0.06)
6.80b
(1.77)
0.29a
(0.04)
3.93a
(2.97)
0.35a
(0.01)
3.40a
(1.25)
0.46b
(0.03)
4.22a
(2.34)
0.55c
(0.04)
4.06a
(1.53)
0.62c
(0.03)
5.54a
(0.64)
109.27a
(7.13)
52.8a
(11.5)
15.4a
(4.7)
139.80a
(10.91)
68.6a
(8.2)
15.5a
(5.3)
206.11b
(21.24)
142.4bc
(17.2)
11.5a
(3.0)
257.08c
(23.06)
116.9b
(19.3)
14.7a
(1.8)
316.86d
(19.84)
163.8c
(18.0)
17.3a
(4.6)
170.20a
(38.75)
81.8a
(16.5)
25.1a
(11.0)
199.54a
(62.15)
93.8a
(14.1)
15.1ab
(3.7)
213.24a
(44.24)
94.4a
(17.3)
18.6a
(6.8)
213.72a
(54.83)
184.6b
(16.9)
9.2b
(4.4)
315.92b
(48.90)
353.4c
(41.3)
6.5b
(1.2)
3.73a
(0.44)
1.01a
(0.12)
4.27a
(0.23)
1.14a
(0.22)
6.55b
(0.31)
2.37b
(0.23)
5.43b
(0.52)
1.54c
(0.31)
7.47b
(0.49)
1.80c
(0.09)
13.55a
(0.81)
2.81a
(0.24)
9.71b
(1.21)
2.67a
(0.36)
9.17b
(0.99)
2.05a
(0.25)
7.38b
(1.21)
3.35b
(0.10)
8.93b
(0.87)
5.70c
(0.91)
For each parameter, different letters indicate significant differences (Tukey HSD Test). Standard deviation is reported in parenthesis.
the C4 sample showed the highest value of Cmic/Corg ×100,
while F and F+C soils showed values very similar to the
untreated soil.
Bacterial counts and molecular characterization
Data concerning the bacterial counts are reported in Table 2
and showed that the number of viable cells ranged between
106 and 107 CFU/g of soil.
Twenty colonies from each plate were further characterized using a combination of molecular PCR-based techniques, with a total of 189 isolates. Firstly, the 16S rRNA
gene was amplified, and each amplicon was treated with
the enzyme AluI as described in “Materials and methods.”
A total of 23 amplified ribosomal DNA restriction analysis
(ARDRA) (Vaneechoutte et al. 1992, 1993, 1995) patterns
were obtained, and some of the restriction profiles appeared
very similar, with many monomorphic bands, thus suggesting a taxonomical closeness of bacterial isolates. Eighteen
of the 23 haplotypes were represented by only a few
isolates (Table 3). The remaining five (A, C, E, N, and Q)
were the main haplotypes, which accounted for about 65%
of the total culturable bacteria.
The nucleotide sequence of the 16S rRNA gene of one
isolate from each ARDRA group was determined. Each of
the sequences obtained was used to probe the nucleotide
databases using the BLASTN option of the BLAST program (Altschul et al. 1997). The most similar orthologous
sequences retrieved were then aligned with the program
ClustalW (Thompson et al. 1994) to the sequences
obtained.
The phylogenetic analysis revealed that the 23 haplotypes
were representative of seven bacterial genera. In particular:
1. Bacteria belonging to 15 different haplotypes (A, B, D,
E, F, I, J, K, M, N, O, Q, R, S, and T) owned to the
genus Bacillus. They represented the largest fraction of
the bacterial community (about 83%). The 15 sequences
felt in nine different clusters and corresponded to nine
different Bacillus species, that is B. firmus, B. lentus, B.
pumilus, B. licheniformis, B. megaterium, B. cereus, B.
simplex, B. endophyticus, and B. psychrodurans.
Table 2 Growth of bacterial colonies on LB medium expressed as n° of bacteria/g fresh soil (data reported are the mean of three replicates)
Month
May
July
Soil treatment
T
F
F+C
C2
C4
2.9±1.1×106
4.2±1.7×106
4.8±1.2×106
7.2±2.3×106
1.6±0.5×107
1.6±0.3×106
1.1±0.2×107
7.8±1.8×106
2.9±1.4×106
4.2±1.5×106
562
Table 3 Distribution of 23 ARDRA haplotypes obtained by restriction analysis of 16S rDNA with the endonuclease AluI from 189 bacteria isolated from different soil samples
ARDRA group
(haplotype)
Soil treatment
T
Number of isolates/
haplotype
F
F+C
% of
total
16S rDNA sequence
determined
C4
C2
July
May
July
May
July
May
July
May
July
A
B
C
D
E
F
8
1
1
0
3
0
6
0
0
0
7
0
7
0
0
0
3
0
1
1
0
1
8
0
9
1
5
0
1
1
7
1
3
0
0
4
1
1
0
0
0
0
5
0
2
3
1
1
1
0
0
0
0
0
4
2
0
1
1
1
49
7
11
5
24
7
26.0
3.7
5.8
2.7
12.7
3.7
G
0
0
0
0
0
0
0
0
3
0
3
1.6
H
I
J
K
L
M
N
O
P
Q
R
S
T
U
V
Z
Number of isolates
analyzed
1
0
1
0
0
0
0
0
0
1
0
0
0
1
1
1
19
0
0
0
0
0
0
0
0
0
6
0
0
0
0
0
0
19
0
0
0
2
1
0
0
0
0
3
0
3
0
0
0
0
19
0
0
1
1
1
0
0
0
0
2
0
2
1
0
0
0
19
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
18
0
0
0
2
0
0
0
0
0
1
0
1
0
0
0
0
19
0
0
0
0
0
0
10
0
2
2
0
0
0
0
0
4
20
2
0
0
3
1
0
0
0
0
1
0
0
0
0
0
0
19
0
2
0
0
0
0
10
2
1
0
0
0
0
0
0
0
19
0
0
0
0
0
2
2
0
2
0
3
0
0
0
0
0
18
3
2
2
9
3
2
22
2
5
16
3
6
1
1
1
5
189
1.6
1.1
1.1
4.7
1.6
1.1
11.6
1.1
2.6
8.4
1.6
3.2
0.5
0.5
0.5
2.6
100
According to 16S rDNA % identity sequence and phylogenetic tree (see also Fig. 1)
B. firmus
B. pumilus
Arthrobacter sp.
B. cereus
B. firmus
B. lentus
B. lentus
Sporosarcina sp.
Sporosarcina sp.
Staphylococcus
B. firmus
B. megaterium
B. simplex
Staphylococcus sp.
B. lentus
B. simplex
B. lentus
Alcaligenes sp.
B. licheniformis
B. psychrodurans
B. endophyticus
B. firmus
Brachybacterium sp.
Pseudomonas sp.
Pseudomonas sp.
Strain
Accession
number
C2L8
F+CM18
F+CM7
C2L18
FL13
F+CL14
F+CL17
C4M11
C4M5
C2L11
C4M15
FL1
C2L3
C2L16
C4M16
C2M7
C4M20
C4M17
FL3
C2M6
FL7
FL20
TM6
TM16
C2M11
DQ073459
DQ073460
DQ073463
DQ073461
DQ073462
DQ112356
DQ112357
DQ112359
DQ112360
DQ076322
DQ089747
DQ089745
DQ089751
DQ089746
DQ089753
DQ073464
DQ112358
DQ089749
DQ089750
DQ089754
DQ089755
DQ089748
DQ089752
DQ078780
–
Biol Fertil Soils (2008) 44:557–569
May
a
Taxonomya
Biol Fertil Soils (2008) 44:557–569
2. Isolates belonging to haplotype C were clustered in the
genus Arthrobacter.
3. Haplotype G joined the genus Sporosarcina.
4. Isolates belonging to haplotypes H and L felt in the
genus Staphylococcus.
5. The five isolates belonging to haplotype P were
clustered in the genus Alcaligenes.
6. Isolate representing haplotype U felt within the genus
Brachybacterium.
7. Isolates belonging to the last two haplotypes (V and Z)
were clustered in the genus Pseudomonas.
The analysis of the data reported above revealed that 20
out of the 23 haplotypes (about 94.2% of the entire
bacterial community studied) were Gram positive with a
prevalence of the genus Bacillus. The alignment of the 16S
rDNA sequences of the genus Bacillus isolates was then
used to construct the phylogenetic tree shown in Fig. 1.
Fluctuations of the bacterial community
The bacterial community was analyzed to check the
existence of a haplotype sharing and to investigate the
effect of two parameters (sampling month and soil
treatment) on variations in bacterial species/genera density
within the community. Data obtained revealed that 19 and
17 haplotypes were found in May and July, respectively
(Fig. 2). Thirteen haplotypes were shared between May and
July samples, whereas six and four haplotypes were
exclusive of May and July samples, respectively.
The distribution of ARDRA haplotypes was analyzed to
check the effect of soil treatment on the dynamics of each
(or group of) ARDRA haplotype. This analysis revealed
that just 1 of the 23 haplotypes, i.e., ARDRA group A (B.
firmus), was represented in all of the samples collected in
May and July (Fig. 3). In addition to this, the total number
of different haplotypes was higher in C (13 in C2 and 12 in
C4) than in T (10), F (10), and F+C (8) soil samples; this
was parallel to the increase in unique haplotypes (7 in
sample C, 2 in T, 1 in F, and 0 in F+C). Therefore, in C
samples, the number of isolates per haplotype was generally
lower than that found in T and F samples. The only
exception was represented by isolates belonging to haplotype N that, according to 16S rDNA analysis, were
assigned to the species B. simplex. Interestingly, all of
these isolates (22), representing more than 28% of the entire
C2 + C4 community, were only found in C samples,
suggesting that the amendment might have positively
affected the growth of these isolates.
Apparently, the increase in haplotype N frequency in
May was parallel to the decrease in the presence of isolates
belonging to ARDRA group E and, although at a lesser
extent, to group A, whose isolates were assigned to the
species B. firmus.
563
Statistical analysis of ARDRA data
To give a statistical support to the ARDRA distribution, the
AMOVA was carried out on ARDRA profiles. AMOVA
data confirmed that most of the total molecular variance
(p=0.001) was ascribed to divergence among strain-specific
haplotypes (91.9%) and in minor part (8.1%) to the
partition of soil types (Table 4). Partitioning the genetic
variance among different groups, a percentage of variation
of 6.9% was ascribed to divergence between group 1
without or with low organic C input (T, F, F+C) and group
2 with high organic inputs (C2, C4).
The AMOVA of ARDRA haplotypes belonging to
Bacillus genus indicated that the molecular variance among
soils increased to 10.6%, and it was probably due to the
divergence among soils (Table 4). Considering the groups 1
and 2 composed only by Bacillus spp., the divergence
increased to 10.2%. B. firmus H haplotype, B. lentus M
haplotype, and B. simplex species present in group 2 and
absent in group 1 showed a high divergence value.
RAPD fingerprinting and statistical analysis of RAPD data
The degree of intraspecific genetic variability between
strains belonging to four main ARDRA groups (A, E, N,
and Q, containing 49, 22, 24, and 16 isolates, respectively)
was analyzed by RAPD fingerprinting (Williams et al.
1990; Welsh and McClelland 1990) as described in
“Materials and methods.”
As expected the amplification patterns of strains belonging to different haplotypes were very different. In addition
to this, a high degree of genetic diversity was found
between isolates belonging to the same ARDRA haplotypes, and 15 RAPD profiles of haplotypes belonging to
group A are shown in Fig. 4. The AMOVA performed on
RAPD profiles of isolates belonging to ARDRA group A
showed that the 91.9% of the total molecular variance was
attributed to divergence among isolates and only 8.0% to
soils. The AMOVA of groups 1 and 2 showed a low value
(6.93%) of divergence.
Discussion
In this work, we have analyzed the effect of fumigant 1,3-D
on the diversity and dynamics of soil heterotrophic aerobic
culturable microbial communities. Although the plate
counts usually estimate only 1–10% of the overall soil microflora, in our opinion, the integration of data concerning
culturable microbial flora and soil parameters may help in
understanding better the effect of fumigants on bacterial
response to changes of environmental parameters. Nevertheless ratios between microbiological parameters have
564
Biol Fertil Soils (2008) 44:557–569
Fig. 1 Phylogenetic tree showing the relationships among the
16S rDNA sequences of the
Bacillus isolates. Scale bars
represent the Kimura 2-parameter distance. Bootstrap values
are indicated at the node. The
accession number of each sequence is also given
often been used for evaluating the response of microbial
activity to environmental factors (Anderson 2003). In this
study, the responses of microbial quotient, qCO2, and qM to
different fertilization treatments, observed during 3 months,
seemed to be strongly affected by the nutritional status of
the soil. It is known that nutrient uptake by microbial cells
is an energetically expensive process, particularly when
microbes are forced to degrade stable organic matter to get
Biol Fertil Soils (2008) 44:557–569
565
30
25
Number of isolates
20
15
10
5
0
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
P
Q
R
S
T
U
V
Z
ARDRA haplotype
Fig. 2 Distribution of ARDRA haplotypes in May (white columns) and July (black columns)
new available substrates. Furthermore, changes in nutrient
availability can modify microbial maintenance energy
requirements. This nutritional “stress” of fumigated soils
could explain the fact that the percentage of organic C
present as microbial biomass C (the microbial quotient) in
May was lower than 2.0, which is considered a critical
threshold for soil health (Anderson 2003). In May, the
highest value of microbial quotient in F+C samples
probably reflects a more efficient use of easily degradable
substrates by microbial biomass (Anderson 2003; Pinzari
et al. 1999).
Data obtained in May showed no significant differences
in chemical and microbiological properties between T and F
samples, indicating that the fertigation alone had no
significant short-time effects on tested soil properties and
microbial activities.
As expected, Corg is a stable parameter, and in May, its
values only significantly changed in soils amended with
high amounts of compost (C2 and C4). The F+C soils
showed the lowest difference between the numbers of
haplotypes from May to July, indicating the maintenance of
the indigenous microbial diversity, although the relative
frequency of each bacterial species changed. Nevertheless,
it is known that, during organic matter mineralization, a
succession of organisms occurs, and this suggests that
different catabolic capabilities are sequentially required to
complete the decomposition process. Most studies on
succession have shown that highly active, zymogenous,
and r-selected organisms dominate the beginning of
decomposition, whereas autochthonous or K-selected, more
energy-efficient organisms occur during later stages (Dilly
et al. 2001). Thus, probably in F+C soils, indigenous rstrategists survived and increased their biomass in May by
mineralizing readily available organic matter and using the
added ammonium nitrate (Dilly 2005). However, the
significant decrease in Cmic in July indicates that the effect
of F+C treatment was just temporary, and it suggests a
decline in organic matter quality and transformation
intensity. Further investigations may provide more data
and reveal fundamental ecological information about the
nutritional physiology of the microbiota, their role in the
mineralization of organic compounds, and their ecophysiological adjustment to environmental conditions.
The highest values of potential mineralization (C0) and
Cmic were observed in soils amended with different amount
of compost (F+C, C2, and C4), probably because the
amendment of soil with compost increased the amount of
degradable organic matter, which may be the main variable
of microbial physiological status; indeed, the qM in July
indicates a stable metabolism in soils treated with compost
while increased in T and F soils probably due to the
depletion of the organic matter. Overall, bacterial counts
were not related to their Cmic values, thus suggesting that
the majority of microbial biomass is formed by uncultur-
566
Biol Fertil Soils (2008) 44:557–569
25
Number of isolates
20
15
10
5
0
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
P
Q
R
S
T
U
V
Z
Haplotype
Fig. 3 Number of strains belonging to the 21 ARDRA haplotypes in relation to soil treatment: T (black columns), F (light gray), F+C (white
columns), C2/C4 (dark gray)
able bacteria and/or by other soil microorganisms such as
fungi or protozoa added with compost.
The values of metabolic quotient were not significantly
different among the different treatments in May; however,
Table 4 AMOVA of 189 bacterial isolates from fumigated soils
Source of
variation
df
Sum of
squares
Variance
components
AMOVA of total ARDRA haplotypes
Among soils
4
7.2
0.04
Within soils
184 76.3
0.41
Total
188 83.4
0.45
Among groups
1
3.9
0.03
Among soils
3
3.2
0.02
within groups
Within soils
184 76.3
0.41
Total
188 83.4
0.46
AMOVA of Bacillus spp. ARDRA haplotypes
Among soils
4
7.3
0.05 Va
Within soils
152 58.5
0.38 Vb
Total
156 65.8
0.43
Among groups
1
4.4
0.05 Va
Among soils
3
2.9
0.02 Vb
within groups
Within soils
152 58.5
0.38 Vc
Total
156 65.8
0.45
Percentage of
variation
8.1
91.9
6.9
3.7
89.4
10.6
89.4
10.3
4.2
85.6
in July, C2 and C4 samples exhibited values significantly
lower than those of T, F, and F+C samples. Increases in
qCO2 could be induced by nutritional stresses as well as by
the increase in the bacterial/fungal ratio (Landi et al. 2000);
it may be possible that the metabolic quotient of T, F, and
F+C soils might have been affected by the low values of
organic matter in July. Furthermore, the increase in qCO2
value in F+C soils in July indicates a short-time effect of
the treatment, and it could be due to the depletion of the
organic C added as compost, as previously discussed.
Both C2 and C4 treatments seem to select specific
microbial species such as B. simplex, which represents
almost the 40% of culturable bacteria isolated in May; this
percentage decreased ten times in July, and B. firmus,
which showed a low number in May, increased 4.5 times in
July, reaching values similar to those of the control. In spite
of fact that the qCO2 value decreased and the microbial
quotient was high probably because mineralization of
organic matter decreased under the stressed conditions in
C2 and C4 samples, the bacterial structure of these samples
in July was similar to that of low-fertilized soils. It is
possible that the addition of high amounts of new organic
matter as compost might have initially favored the growth
of less-represented bacterial species such as B. simplex;
afterwards, the microbial community recovered the preexisting equilibrium. However, we cannot a priori rule com-
Biol Fertil Soils (2008) 44:557–569
Fig. 4 a Example of RAPD
patterns of total DNA with
primer AP12 from isolates of
ARDRA haplotype A. b Cluster
analysis based on UPGMA of
RAPD profiles. Scale bar
numbers indicate divergences
among profiles
567
a
A1 A2 A3
A4
A5 A6
A7 A8 A9 A10 A11 A12
A13 A14 A15
b
A6
A9
A8
A1
A5
A7
A12
A14
A3
A15
A10
A11
A4
A2
A13
pletely out the possibility that the addiction of compost
might have introduced not detectable and/or less-represented
microbial species (e.g., B. simplex), which resulted in an
initially change in the composition of soil microflora.
The AMOVA performed on ARDRA haplotypes belonging to Bacillus genus showed that the divergence
between groups 1 (without or very low fertilization rate)
and 2 (with high fertilization rate) was 10.2%, thus
indicating a different distribution of Bacillus spp. strains
among soils.
On the basis of molecular content in G+C, phenotypic
similarity, and 16S rDNA group, Priest (1993) subdivided
568
the genus Bacillus into five different groups. The phylogenetic tree reported in Fig. 1 revealed that a very high
percentage of the Bacillus isolates (more than 94%) were
placed within group 2, which is mostly composed by
species which are facultative anaerobes, grow strongly in
absence of oxygen, produce ellipsoidal endospores, and
swell the mother cell (Priest 1993). In particular, B. firmus
and related halotolerant bacteria such as B. lentus are
particularly common in marine and estuarine habitats and
salt marshes (Gordon and Hyde 1982; Gordon et al. 1977).
Alkaliphilic bacilli are widespread in soils, being found in
acidic soils but being more common in alkaline soils like the
one analyzed in this work. Therefore, it is possible that the
selective pressure induced by 1,3-D had strongly favored
microorganisms resistant to the fumigant through the
formation of spores. However, we cannot a priori exclude
the possibility that a number of strains might have survived
by degrading the fumigant agents. If this idea is correct, the
RAPD analysis should have revealed a selection of particular
strain(s). The findings of a high degree of genetic variability
between strains belonging to the main ARDRA groups and
the fact that this biodiversity was not influenced by the soil
treatments as shown by AMOVA contradicts what was
previously reported by Gan et al. (1998), that the fumigants
select at the species rather than at the strain level. Thus, it is
quite possible that the presence of a high percentage of
Gram-positive bacteria in the fumigated soil, according to
Ibekwe et al. (2001a), and in particular, the prevalence of the
genus Bacillus might be related to the ability of these
bacteria to form spores to protect themselves from the
fumigants rather than to the presence of a set of genes
involved in biodegradation of 1,3-D.
In conclusion, the amendment of fumigated soils is a
temporary solution to recover the loss of soil fertility by
1,3-D, which selects the bacterial community. However,
further researches, such as analysis of the unculturable
microflora, the analysis of composition of compost bacterial
communities, and/or longer-term analysis of the effect of
the compost addiction, are required to improve one of these
treatments as a possible solution for recovering the fertility
of the fumigated soils.
Acknowledgment We are very grateful to two anonymous
reviewers and to the Editor-in-chief for the revision of the manuscript
and for their useful suggestions and comments in improving the
manuscript.
References
Altschul SF, Madden TL, Schäffer AA, Zhang J, Zhang Z, Miller W,
Lipman DJ (1997) Gapped BLAST and PSI-BLAST: a new
generation of protein database search programs. Nucl Acids Res
25:3389–3402
Biol Fertil Soils (2008) 44:557–569
Anderson TH (2003) Microbial eco-physiological indicators to assess
soil quality. Agric Ecosyst Environ 98:285–293
Anderson TH, Domsch KH (1990) Application of eco-physiological
quotients (qCO2 and qD) on microbial biomass from soils of
different cropping histories. Soil Biol Biochem 10:251–255
Baldwin JG, Nadler SA, Adams BJ (2004) Evolution of plant
parasitism among nematodes. Annu Rev Phytopathol 42:83–105
Brookes PC (1995) The use of microbial parameters in monitoring soil
pollution by heavy metals. Biol Fertil Soils 19:269–279
Burns RG, Nannipieri P, Benedetti A, Hopkins DW (2006) Defining
soil quality. In: Bloem J, Hopkins DW, Benedetti A (eds)
Microbiological methods for assessing soil quality. CABI
publishing, pp 15–22
Dilly O (2005) Microbial energetics in soils. In: Buscot F, Varma A
(eds) Microrganisms in soils: roles in genesis and functions.
Springer, Berlin, pp 123–128
Dilly O, Munch JC (1998) Rations between estimates of microbial
biomass content and microbial activity in soils. Biol Fertil Soils
27:374–379
Dilly O, Bartsch S, Rosenbrock P, Buscot F, Munch JC (2001) Shifts
in physiological capabilities of the microbiota during the
decomposition of leaf litter in a black alder (Alnus glutinosa
(Gaertn.) L.) forest. Soil Biol Biochem 33:921–930
Doran JW, Parkin TB (1994) Defining and assessing soil quality. In:
Doran JW, Coleman DC, Bezdicek DF, Stewart BA (eds)
Defining soil quality for a sustainable environment. Soil Science
Society of America, Madison, Wisconsin, pp 3–22 (Special
Publication no.35 SSSA)
Excoffier L, Smouse PE, Quattro M (1992) Analysis of molecular
variance inferred from metric distances among DNA haplotypes:
application to human mitochondrial DNA restriction data.
Genetics 131:479–491
Gan J, Yates SR, Crowley D, Becker JO (1998) Acceleration of 1,3dichloropropene degrading by organic amendments and potential
application for emission reduction. J Environ Qual 27:408–414
Giannakou IO, Sidiropoulos A, Prophetou-Athanasiadou D (2002)
Chemical alternatives to methyl bromide for the control of rootknot nematodes in greenhouses. Pest Manag Sci 58:290–296
Gordon RE, Hyde JL (1982) The Bacillus firmus–Bacillus lentus
complex and pH 7.0 variants of some alkalophilic strains. J Gen
Microbiol 128:1109–1116
Gordon RE, Hyde JL, Moore JA Jr (1977) Bacillus firmus. –Bacillus
lentus: a series or one species. Int J Syst Bacteriol 27:256–262
Grifoni A, Bazzicalupo M, Di Serio C, Fancelli S, Fani R (1995)
Identification of Azospirillum strains by restriction fragment
length polymorphism of the 16S rDNA and of the histidine
operon. FEMS Microbiol Lett 127:85–91
Ibekwe AM, Papiernik SK, Gan J, Yates SR, Yang CH, Crowley DE
(2001a) Impact of fumigants on soil microbial communities.
Appl Environ Microbiol 67:3245–3257
Ibekwe AM, Papiernik SK, Gan J, Yates SR, Yang CH, Crowley DE
(2001b) Microcosm enrichment of 1,3-dichloropropene-degrading
soil microbial communities in a compost-amended soil. J App
Microbiol 91:668–676
Isermeyer H (1952) Eine Einfache Methode sur Bestimmung der
Bodenatmung und der Karbonate im Boden. Z Pflanzanernah
Bodenk 56:6–38
Jenkinson DS, Ladd JN (1981) Microbial biomass in soil: measurements and turnover. In: Paul EA, Ladd JN (eds) Soil biochemistry, vol. 5. Marcel Dekker, New York, pp 415–471
Kimura M (1980) A simple method for estimating evolutionary rate of
base substitutions through comparative studies of nucleotide
sequences. J Mol Evol 16:11–120
Kumar S, Tamura K, Nei M (2004) MEGA3: integrated software for
molecular evolutionary genetics analysis and sequence alignment. Brief Bioinf 5:150–163
Biol Fertil Soils (2008) 44:557–569
Landi L, Renella G, Moreno JL, Falchini L, Nannipieri P (2000)
Influence of cadmium on the metabolic quotient, L-:D-glutamic
acid respiration ration and enzyme activity: microbial biomass
ratio under laboratory conditions. Biol Fertil Soils 32:8–16
Ministero delle Politiche Agricole e Forestali (1997) Metodi ufficiali
di analisi fisica del suolo. Decreto Ministeriale 1° agosto 1997.
In: Supplemento ordinario alla Gazzetta Ufficiale n°204 del 2
Settembre 1997
Ministero delle Politiche Agricole e Forestali (1999) Metodi ufficiali
di analisi chimica del suolo. Decreto Ministeriale 13 settembre
1999. In: Supplemento ordinario alla Gazzetta Ufficiale n°248
del 21 Ottobre 1999
Mori E, Lio’ P, Daly S, Damiani G, Perito B, Fani R (1999) Molecular
nature of RAPD markers amplified from Haemophilus influenzae
rd genome. Res Microbiol 150:83–93
Nannipieri P, Grego S, Ceccanti B (1990) Ecological significance of
the biological activity in soil. In: Bollag JM, Stotzky G (eds) Soil
biochemistry, vol 6. Marcel Dekker, New York, pp 293–356
Ou LT, Chung KY, Thomas JE, Obreza TA, Dickson DW (1995)
Degradation of 1,3-dichloropropene (1,3-D) in soils with different histories of field applications of 1,3-D. J Nematol 27:127–
242
Park MK, Kim JH, Dungan RS (2004) Sorption of the fumigant 1,3dichloropropene on soil. J Environ Sci Health B 39:603–612
Pinzari F, Trinchera A, Benedetti A, Sequi P (1999) Use of
biochemical indices in the Mediterranean environment: comparison among soils under different forest vegetation. J Microbiol
Meth 36:21–28
Prather MJ, McElroy MR, Wofsy SC (1984) Reduction in ozone at
high concentrations of stratospheric halogens. Nature 31:227–
231
Priest FG (1993) Systematics and ecology of Bacillus. In: Hoch JA,
Losick R (eds) Bacillus subtilis and other Gram-positive bacteria,
systematics and ecology of Bacillus. American Society for
Microbiology, Washington, DC, pp 3–16
Riffaldi R, Saviozzi A, Levi-Minzi R (1996) Carbon mineralization
kinetics as influenced by soil properties. Biol Fertil Soil 22:293–
298
Saitou N, Nei M (1987) The neighbor-joining method: a new method
for reconstructing phylogenetic trees. Mol Biol Evol 4:406–425
569
Sakamoto K, Oba Y (1994) Effect of fungal to bacterial biomass ratio
on the relationship between CO2 evolution and total soil
microbial biomass. Biol Fertil Soil 17:39–44
Sanger F, Nicklesen S, Coulson AR (1977) DNA sequencing with chain
terminating inhibitors. Proc Natl Acad Sci U S A 74:5463–5467
Schneider S, JM Kueffer, D Roessli, L Excoffier (1997) ARLEQUIN:
a software for population genetics data analysis. Version 1.1,
University of Geneva
Smelt JH, Teunissen W, Crum SJH, Leistra M (1989) Accelerated
transformation of 1,3-dichloropropene in loamy soils. Neth J
Agric Sci 37:173–183
Springer U, Klee J (1954) Prüfung der Leistungsfähigkeit von einigen
wichtigeren Verfahren zur Bestimmung des Kohlemstoffs mittels
Chromschwefelsäure sowie Vorschlag einer neuen Schnellmethode. Z Pflanzenernähr Dang Bodenk 64:1
Thompson JD, Higgins DG, Gibson TJ (1994) CLUSTAL W:
improving the sensitivity of progressive multiple sequence
alignment through sequence weighting, position-specific gap
penalties and weight matrix choice. Nucl Acids Res 22:4673–
4680
Vance ED, Brookes PC, Jenkinson DS (1987) An extraction method
for measuring soil microbial biomass C. Soil Biol Biochem
19:703–707
Vaneechoutte M, Rossau R, De Vos P, Gillis M, Janssens D, Paepe N
(1992) Rapid. identification of the bacteria of the Comamonadaceae with amplified ribosomal DNA-restriction analysis
(ARDRA). FEMS Microbiol Lett 93:227–234
Vaneechoutte M, De Beenhouwer H, Claeys G, Verschraegen GM, De
Rouck A, Paepe N (1993) Identification. of Mycobacterium
species with amplified rDNA restriction analysis. J Clin Microbiol 31:2061–2065
Vaneechoutte M, Dijkshoorn L, Tjernberg A, Eilachouni A, De Vos P,
Claeys G, Verschraegen G (1995) Identification. of Acinetobacter genomic species by amplified ribosomal DNA restriction
analysis. J Clin Microbiol 33:11–15
Welsh J, McClelland M (1990) Fingerprinting genomes using PCR
with arbitrary primers. Nucl Acids Res 18:7213–7218
Williams JGK, Kubelik AR, Livak KJ, Rafalski JA, Tingey SV (1990)
DNA polymorphisms amplified by arbitrary primers are useful as
genetic markers. Nucl Acids Res 18:6531–6535