1 J...
H..wr........m&.%
. .i.
...:..u... virui
(Slivmnnuhrh
‘4‘“: ‘5
. ..

 

.hrlinzf . ‘

.

 

..
J

.nVurL»

a: .

a
fiwfimjfii
Elgtfl: $de5 .
‘ n v I. .
. , 9. i...

SEATIGXSp I

.liszJAn l

dung. in...

$1:

5;; :25... I: A :1:

via-.11. xvx..1h)£h.t.ll.:
Quunt VA...

 

 

 

4

. ‘ ‘ K. 1:4?!

1.» i , u (7:1:

339‘ .. an . . 3
15.5.! i: .

3.! -
#532 A,“
:3 2,1 ....

   

 

n;

3km“

, .23.. {-29
5:19;. N» v .21

. . : x £31494“... 1?. 55".”? to? .1

....d:;...r.....; “5% . 35755. . 3.. . u

I d is: 33:; : , . . . . . . Wink L!»

i.r’.\$
(I...

 

:HESlS

IlllflllllJill?lllllllllllllllllll

1293 01567 3977

 

This is to certify that the

dissertation entitled

NUCLEIC ACID ANALYSIS OF AUTOTROPHIC AMMONIA-
OXIDIZING BACTERIA IN SOILS

presented by

MARY ANN BRUNS

has been accepted towards fulfillment
of the requirements for l

Ph.D. degreein Crop and Soil Sciences

fizz/M6

Major professor

/ ,_
Dane/éL/LM

MS U is an Affirmative Action/Equal Opportunity Institution 0-12771

 

 

LIBRARY

Michigan State
University

 

 

 

PLACE III RETURN BOX to romovo this chookout from your rooord.
TO AVOID FINES rotum on or bdoro duo duo.

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU to An Afflrmotivo Action/Equal Opportunity Institution
WM!

_.. __———--u————-—.—.

NUCLEIC ACID ANALYSIS OF AUTOTROPHIC AMMONIA-OXIDIZING
BACTERIA IN SOILS

By
Mary Ann Bruns

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Crop and Soil Sciences

1996

ABSTRACT

NUCLEIC ACID ANALYSIS OF AUTOTROPHIC AMMONIA-OXIDIZING
BACTERIA IN SOILS

By
Mary Ann Bruns

Nitrification, the microbial oxidation of ammonium to nitrate, leads to
significant nitrogen (N) losses from soils. Nitrification rates in soils have been
related to population sizes of ammonia-oxidizing bacteria, but population
structure effects are poorly understood. This study employed nucleic acid-
based methods to determine how ammonia oxidizer populations differed in
cultivated, never-tilled, and successional soils. Hybridizations were
conducted between genomic DNA from pure cultures and probes from
Nifrosomonas europaea genes for ammonia monooxygenase and
hydroxylamine oxidoreductase. These probes produced 24 to 80% less
hybridization signal in genomic DNA from four other ammonia oxidizers under
low-stringency conditions. This indicated that N. europaea functional probes
would not provide quantitative size estimates for heterogeneous populations.
Sizes of ammonia oxidizer populations in soils were determined by Most
Probable Number (MPN) enumeration in media containing 10, 100, or 2000
ppm NH4-N. MPN counts were consistently one-tenth lower in never-tilled
soils than in cultivated soils. Population structures were compared by using
PCR to amplify 16S rFtNA genes of ammonia oxidizers from bacterial
community DNA, cloning the PCR products, and comparing cloned sequences
for percent similarity over a 318-base region. The 168 rRNA gene sequences

obtained from never-tilled soils were four times more diverse than sequences

obtained from cultivated soils. A phylogenetic tree showed that sequences
from cultivated soils fell within one distinct cluster of Nitrosospira; sequences
from never-tilled soils fell into two clusters of Nifrosospira and one of
Nitrosomonas. Denaturing gradient gel electrophoresis (DGGE) patterns
differed between samples from never-tilled and cultivated soils; patterns from
samples from replicate plots of each treatment were very similar in each of two
successive years. Probe hybridizations of DGGE blots showed that
sequences from the single subgroup of Nitrosospira spp. predominated in
samples from cultivated or fertilized successional soils; these were not
detected in samples from never-tilled or unfertilized successional soils.
Differences in populations were reflected in mean NH4-N concentrations that
were tenfold higher in never-tilled soils than in cultivated soils. Different
influential factors in these soils include pH, allelopathy, microsite distribution,

or direct effects of cultivation and fertilization.

DEDICATION

This dissertation is dedicated to the Center for Rural Affairs, a private,
nonprofit, rural advocacy organization in Walthill, Nebraska. The Center for
Rural Affairs' research and education efforts have given me professional

guidance which I hope has shaped my perspective as a soil scientist.

iv

ACKNOWLEDGMENTS

The Center for Microbial Ecology (CME) and the Department of Crop
and Soil Sciences (083) at Michigan State University (MSU) have provided a
highly congenial and cooperative environment for graduate research. I have
greatly appreciated the opportunities provided at MSU and am especially
grateful to Dr. Eldor Paul, my major adviser, for his encouragement and
direction. I would like to thank my graduate committee members, Dr. Sharon
Anderson, Dr. Richard Harwood, Dr. Bill Holben, and Dr. Mike Klug, for all of
their helpful advice these past six years. I am also indebted to two other CSS
faculty members, Dr. Jim Tiedje and Dr. Phil Robertson, who have given me
key support and contacts for my research. Dr. Tiedje and Dr. Robertson,
respectively, direct the CME and Long-Term Ecological Research (LTER)
projects, without which this dissertation project would not have been possible.
I also wish to thank Dr. Jim Presser, University of Aberdeen. Scotland, for
making my research in Aberdeen possible.

Many, many people have given me help and support. From the Kellogg
Biological Station, I would especially like to thank Sandy Halstead, for help
with sampling, Martha Tomecek and Carolyn Miller, for help with data
provision, and Hal Collins for his advice. Persons in the CME who have given
me valuable technical support are Dave Harris and Joyce Wildenthal. Other

colleagues in the CME who have been generous with their time, friendship,

and advice are Olga Maltseva, Craig Meyer, Klaus Nuesslein, Jim Cole, John
Urbance, Frank Loeffler, Michel Cavigelli, Grace Matheson, Bonnie Bratina,
Tamara Tsoi, Shannon Flynn, and my 'labmates' Francisco Calderon, Chris
Wright, and John Ravenscroft. I also want to thank CME 'alumni" Cathy
McGowan, Alenka Princic, Jizhong Zhou, Marcos Fries, Roberta Fulthorpe,
Cindy Nakatsu, Sheridan Haack, Jim Champine, Suzanne Theim, Jenny
Norton, Rob Sanford, and Joanne Chee-Sanford for their support and
friendship. Yuichi Suwa also gave me enthusiastic advice and mentoring for
working with nitrifiers.

The CME staff, Brenda Minott, Betty Fancher, Lisa Pline, and Evelyn
Yaeger, have never failed to provide me with everything I needed in the office.
I will miss the pleasant working environment that they have provided for us
students. In 088, I would like to thank Elaine Parker, Geri Wardwell, Anne
Conwell, and Mark Bitman for all of their help as well. Jackie Wood, at MSU's
Center for Electron Microscopy, did beautiful work on the transmission
electron micrographs for this dissertation. Wei-Man Wu, at the Michigan
Biotechnology Institute, provided methanotrcph samples that should provide
the basis for future evaluation of the probes used in this project.

I am grateful to Dr. Dan Arp and Dr. Norman Hommes, of Oregon State
University, who provided me with clones for functional probes, and to Dr. Ed
Schmidt, of the University of Minnesota, who provided ammonia oxidizer
cultures. Dr. George Kowalchuk, at the Netherlands Institute of Ecology, gave
valuable help with sample analysis. And last but not least, I would like to
recognize my colleagues in Dr. Prosser's lab at the University of Aberdeen in
Scotland--John Stephen (for his mentoring on molecular techniques and
advice on the Scottish dialect), Zena Smith (for her help and friendship), and
Allison McCaig (for the primers that got it all started).

vi

TABLE OF CONTENTS

LIST OF TABLES ................................................................................................ x

LIST OF FIGURES ............................................................................................ xiii
CHAEIZEBQNE

OVERVIEW ......................................................................................................... 1

REFERENCES ....................................................................................... 12
QUAELEBM

CULTURAL METHODS FOR AMMONIA-OXIDIZING BACTERIA .................. 15

INTRODUCTION .................................................................................... 15

MATERIALS AND METHODS ................................................................ 16

Strains .......................................................................................... 16

Liquid culture media ..................................................................... 17

Solid medium ............................................................................... 19

Treatments and soils .................................................................... 19

Most-probable-number (MPN) analyses ...................................... 21

Enrichments from never-tilled soils .............................................. 24

Long-term storage methods for AAOs ......................................... 24

RESULTS ............................................................................................... 25

AAO growth on solid medium ...................................................... 25

Most-Probable-Number (MPN) analyses ..................................... 26

Enrichments ................................................................................. 34

Long-term storage of MOS ......................................................... 34

DISCUSSION ......................................................................................... 36

REFERENCES ....................................................................................... 45

CHAEIEBIHBEE
EVALUATION OF NIT ROSOMONAS EUROPAEA GENE PROBE

HYB RIDIZATIONS WITH NIT ROSOSPIRA DNA ............................................. 47
INTRODUCTION .................................................................................... 47
MATERIALS AND METHODS ................................................................ 50

DNA probes .................................................................................. 50
Bacterial strains ............................................................................ 50
AAO batch culture scale-up, DNA extraction, and REP-PCR ...... 53
Restriction digests, Southern blots, and probe hybridizations ..... 54
Densitometric measurements from autcradiograms .................... 56

vii

RESULTS ............................................................................................... 57

Genomic DNA yields from AAO cultures ...................................... 57
REP-PCR patterns ....................................................................... 57
DNA probe hybridization patterns ................................................ 57
Comparisons of summed band intensities ................................... 63
Hybridization results with other DNA probes ................................ 66
DISCUSSION .......................................................................................... 66
ACKNOWLEDGEMENTS ....................................................................... 70
REFERENCES ....................................................................................... 71

W
USE OF 165 rDNA ANALYSIS TO COMPARE DIVERSITY OF

AUTOTROPHIC AMMONIA OXIDIZER POPULATIONS IN SOILS ................ 75
INTRODUCTION .................................................................................... 75
MATERIALS AND METHODS ................................................................ 77

Long-Term Ecological Research site ........................................... 77
Soils and treatments .................................................................... 77
DNA extraction from soils ............................................................. 81
Polymerase chain reaction with 16S rDNA primers ..................... 81
Cloning of PCR products .............................................................. 82
T-tracking and sequencing of PCR inserts ................................... 83
Colony blots and probe hybridizations ......................................... 84
Denaturing gradient gel electrophoresis (DGGE) ........................ 85
RESULTS ............................................................................................... 87
DNA yields from soils ................................................................... 87
Sequence diversity in clone libraries ............................................ 89
Probe hybridizations of clone-blots .............................................. 93
Denaturing gradient gel electrophoresis (DGG E) ........................ 95
DISCUSSION .......................................................................................... 99
REFERENCES ..................................................................................... 1 1 1

AEEENQILA

DNA RECOVERY FROM SOILS OF DIVERSE COMPOSITION ................... 114
INTRODUCTION .................................................................................. 1 15
MATERIALS AND METHODS .............................................................. 115

Soils ........................................................................................... 115
Bacterial strains and soil inoculation .......................................... 115
Effect of CTAB and PVPP on humic contamination of

crude extracts .......................................................................... 115
SDS-based DNA extraction methods ......................................... 116
Cell lysis and direct microscopic counts ..................................... 116
DNA extraction from gram-positive bacteria .............................. 116
Purification of crude DNA extracts ............................................. 116
DNA quantification ..................................................................... 117
PCR, restriction enzyme digestion, and Southern blotting ......... 117

viii

RESULTS ............................................................................................. 1 17

Soil properties ............................................................................ 1 17
Effect of CTAB or PVPP on DNA extraction from NK soil .......... 117
Evaluation of DNA extraction and cell lysis on more
challenging soils ...................................................................... 117
Comparison of different methods for lysing
gram-positive bacteria ............................................................. 118
Comparison of DNA purification methods with crude DNA
from NK soil ............................................................................. 118
Evaluation of purification methods on crude DNA extracts
from more challenging soils .................................................... 119
DISCUSSION ........................................................................................ 120
ACKNOWLEDGMENTS ....................................................................... 121
REFERENCES ..................................................................................... 121

ix

LIST OF TABLES

TABLE ......................................................................................................... PAGE
Chapter Two

TABLE 2.1. Compositions of modified ATCC Medium 929 and ATCC

Medium 1573 ..................................................................................................... 18

TABLE 2.2. NH4-N concentrations used in MPN recovery media ................... 23

TABLE 2.3. Comparative MPN counts (per gram of soil) obtained in Schmidt-
Belser medium with and without bentonite clay (1992 soil samples) ................. 31

TABLE 2.4. Comparative MPN counts (per gram dry soil) obtained in Medium
929 containing 1/10X, 1X, and 10X standard NH4-N concentrations (1993 soil
samples) ............................................................................................................ 32

TABLE 2.5. Comparative MPN counts obtained in Medium 929 containing
1/10X, 1X, and 20X standard NH4-N concentrations (1995 data) ..................... 33

TABLE 2.6. Evaluation of effect of freezing and frozen storage on viability of
MOS: number of days required for cultures to turn yellow after thawing and
transfer to fresh medium .................................................................................... 35

TABLE 2.7. Inorganic N measurements (N03-N plus NH4-N) during 1993 and
1994 growing seasons in cultivated, never-tilled, and unfertilized successional
soils, with n=30 .................................................................................................. 42

Chapter Three

TABLE 3.1. DNA probes used in this study ..................................................... 51

TABLE 3.2. Strains of autotrophic ammonia-oxidizing bacteria used in
this study ............................................................................................................ 52

TABLE 3.3. Summed intensities of all hybridization bands obtained with
each probe from genomic DNA of five ammonia oxidizer strains ...................... 64

TABLE 3.4. Summed band intensities for N. multifonnis, Nifrosolobus sp. strain
24-C, Nifrosospira sp. strain AV, and Nifrosospira sp. strain 3919, expressed
as percentages of the summed intensities for N. europaea (Table 2.3) ............ 65

Chapter Four

TABLE 4.1. Management histories, biochemical measurements, and biological
characteristics of soils sampled from cultivated, successional, and never-tilled
treatments at the LTER site, Kellogg Biological Station, Michigan, U. S. A.
(includes 1993, 1994, and 1995 data) ............................................................... 80

TABLE 4.2. Oligonuclectide probe sequences identified from the database of

110 partial rDNA sequences from B—subgroup autotrophic ammonia-oxidizing
bacteria (Stephen et al., 1996a) ........................................................................ 86

TABLE 4.3. DNA yields (ug DNA per 9 soil dry weight) from cultivated, never-
tilled, and successional soils .............................................................................. 88

TABLE 4.4. Numbers of different partial 16$ rDNA sequences falling within
subgroups in B-Proteobacteria (1994 samples) ................................................. 94

TABLE 4.5. Percentages of probe-positive clones in libraries obtained from soil
community DNA (1994 samples) ....................................................................... 96

TABLE 4.6. Numbers of probe-positive clones in libraries obtained from soil
community DNA (1995 samples) ....................................................................... 97

TABLE 4.7. Indicators of the effects autotrophic ammonia oxidizers may have
on nitrogen cycle parameters, measured in two replicate plots of cultivated,
successional, and never-tilled treatments (1994 LTER data) .......................... 108

Appendix A

TABLE A.1. Sampling locations and properties of soils used in DNA extraction
procedures ....................................................................................................... 1 16

TABLE A.2. Comparison of DNA yield and purity of the crude DNA from
seeded NK soil subjected to different treatments ............................................ 117

TABLE A.3. Crude DNA yields from eight soils and percentages of crude DNA
recovered in the large-scale purification method involving gel electrophoresis

and passage through one Megacolumn with 10 ml of Wizard Minipreps plasmid
purification resin ............................................................................................... 11

TABLE A.4. Direct counts and DNA yields for individual samples of eight
soils .................................................................................................................. 118

TABLE A.5. Comparison of DNA yield from gram-positive bacteria by different
lysis methods ................................................................................................... 11 9

TABLE A.6. DNA recovery for different purification methods and enzyme
digestion of the DNA samples purified from unseeded NK soil ....................... 119

xii

LIST OF FIGURES

 

FIGURE PAGE
Chapter One

FIGURE 1.1. Structure of 16S rRNA molecule of Escherichia coli ..................... 6
Chapter Two

FIGURE 2.1. Transmission electron microscopy photograph of Nitrosomonas
europaea ATCC 25978 showing intracytoplasmic membranes as Iamellae at the
periphery of the cell (39,000X magnification) .................................................... 27

FIGURE 2.2. Transmission electron microscopy photograph of Nitrosolobus
multiforrnis ATCC 25196 showing intracytoplasmic membranes
compartmentalizing cell into a lobate structure (29,000 X magnification) .......... 28

FIGURE 2.3. Transmission electron microscopy photograph of Nitrosospira
strain Np AV (Schmidt) showing lack of intracytoplasmic membranes; cell is
comma-shaped, due to a single spiral turn (39,000X magnification) ................. 29

FIGURE 2.4. Transmission electron microscopy photograph of Nifrosospira
strain R1 grown on Noble agar plates containing mineral medium and CaCOa

(29,000X magnification) ..................................................................................... 30

FIGURE 2.5. Mean NOa-N levels in cultivated (T1), never-tilled (T8), and

successional (17) soils in 1993 and 1994 .......................................................... 38

FIGURE 2.6. Mean NH4-N levels in cultivated (T1), uncultivated (T8) and

successional soils (T 7) in 1993 and 1994 ......................................................... 39
Chapter Three

FIGURE 3.1. Ethidium-bromide-stained agarose gel containing EcoRl-digested
DNA of ammonia oxidizers and control bacteria ................................................ 55

FIGURE 3.2. Rep-PCR patterns of autotrophic ammonia-oxidizer DNA
observed in agarose gels stained with ethidium bromide and illuminated under
UV light .............................................................................................................. 58

FIGURE 3.3. Autoradiogram photographs from Southern blots of genomic DNA
hybridized with amoA probe .............................................................................. 59

xiii

FIGURE 3.4. Autoradiogram photographs from Southern blots of genomic DNA

hybridized with hao probe .................................................................................. 60
FIGURE 3.5. Autoradiogram photographs from Southern blots of genomic DNA
hybridized with hcy probe .................................................................................. 61
FIGURE 3.6. Southern blots of AAO genomic DNA hybridized with amoA

probe (left) and amoB probe (right) ................................................................... 62

FIGURE 3.7. Autoradiogram photograph of Southern blot hybridized with nir
probe éor a copper-type nitrite reductase gene obtained from Pseudomonas sp.
strain -179 ....................................................................................................... 67

Chapter Four

FIGURE 4.1. T-track banding patterns observed with single-dideoxynucleotide
termination sequencing reactions ...................................................................... 90

Flgure 4.2. Phylogenetic tree showing Euclidian distances between 168 rDNA
sequences obtained from cultivated soil samples (KZOO-H sequences) and
never-tilled soil samples (KZOO-D sequences) ................................................. 92

FIGURE 4.3. DGGE band patterns of PCR-amplification products from
community DNA observed in ethidium-bromide-strained denaturing gradient

gels under UV illumination ................................................................................. 98

FIGURE 4.4. Autoradiogram of DGGE blot probed with 32P-Iabelled

Spira__Cl_3 probe ............................................................................................. 100

FIGURE 4.5. Autoradiogram of DGGE blot probed with 32P-iabeued All-Spira

probe ................................................................................................................ 101
Appendlx A

FIGURE A.1. Agarose gel electrophoresis of total DNA extracted from NK soil
by different treatments ..................................................................................... 117

FIGURE A.2. Agarose gel electrophoresis of 168 rDNA amplification products
from DNA of unseeded NK soil samples ......................................................... 119

FIGURE A.3. Autoradiogram of hybridization signals with the 0ch gene after
Southern transfer of DNA from seeded sterile NK soil .................................... 120

FIGURE A.4. Agarose gel electrophoresis of 168 rDNA amplification products
of DNA samples from six soils purified by the large-scale procedure .............. 120

xiv

Chapter One
OVERVIEW

Nitrification, the microbial oxidation of ammonia to nitrate, can lead to
significant nitrogen (N) losses from ecosystems. Nitrification occurs as a two-
step process carried out by two different groups of autotrophic bacteria:
ammonia-oxidizing bacteria first convert ammonia to nitrite; then nitrite oxidizers
convert nitrite to nitrate. Both groups oxidize inorganic N as their principal
source of energy and use 002 as their principal source of C (Book at al., 1989).
Autotrophic nitrifiers must be distinguished from heterotrophic nitrifiers, which
include some soil fungi and soil bacteria such as Arthrobacfer spp. (Kurokawa
et al., 1985). The heterotrophic nitrifiers oxidize reduced inorganic N in
cometabolic reactions, but not as a means of generating energy (Killham,
1986). Heterotrophic nitrification constitutes only a small fraction of total
nitrification in most soils (Kuenen and Robertson, 1988).

Virtually all soils containing NH4-N are subject to nitrification, and
nitrification rates vary widely. Nitrification rates are typically low in 'undisturbed
soils, such as grasslands and some forest soils, compared to cultivated soils
Miich receive N fertilizer and undergo disturbance (Clark and Paul, 1970;
Schmidt and Belser, 1982). Nitrification rates are controlled principally by NH4-
N concentration, but other environmental factors such as pH, temperature,
oxygen content, and moisture also play important roles (Robertson, 1990). The
size of the autotrophic nitrifier population is related to nitrification activity in soils
(Belser and Mays, 1982), but the influence of population structure on nitrification

rates is poorly understood.

2

The autotrophic nitrifiers represents a critical link in N cycling within
microbial and plant communities. In acid and neutral soils nearly all ammonia
is protonated. When N is in the form of ammonium (NH4+), it is retained in the
soil as an exchangeable cation bound to negatively charged sites on mineral or
organic soil components (Young and Aldag, 1982). Ammonium can also be
retained as a nonexchangeable cation held in clay mineral interlayers or
organic matter complexes (Nornmik and Vahtras, 1982). In the intracellular
environment of ammonia-oxidizing bacteria, the substrate that is oxidized is
ammonia. When nitrifying bacteria convert ammonia to nitrate, N in the anion
form is no longer held at negatively charged sites on soil colloids. Nitrate is
mobile and susceptible to loss from the system by leaching into groundwater or
by denitrification as nitrogen gases into the atmosphere. The activity of nitrifying
bacteria is a major factor in determining whether the soil retains or loses
nitrogen. Ammonia oxidizers are the more functionally important group of
nitrifying bacteria, since they carry out the first, rate-determining step in
nitrification (Koops and Moller, 1992).

The objective of this study was to determine whether soils with different
disturbance histories select for different populations of autotrophic ammonia
oxidizers (AAOs). Soils for this study were obtained from replicated, 1-ha plots
at the Long-Term Ecological Research (LTER) site at Kellogg Biological Station
near Kalamazoo, Michigan. This site was established in 1989 to study
ecological interactions, nutrient availability, and biotic diversity of cropped,
native, and successional ecosystems that are representative of the Upper
Midwest USA (Robertson et al., 1996). The two treatments with the most
distinctive disturbance histories at LTER are Treatment 1 (high-input,
conventionally tilled rotation of corn and soybeans) and Treatment 8 (never-

tilled, native successional). Prior to 1989, Treatment 1 plots had been cropped

3

in corn and soybeans for at least 40 years. These plots have been moldboard
plowed, disked, and cultivated every year since 1989, and they have been more
intensively fertilized than all other treatments at LTER. Treatment 8 plots, which
were cleared of native deciduous forest in 1959, have never been cropped or
fertilized. A third treatment, which has a disturbance history intermediate
between Treatment 1 and Treatment 8, is Treatment 7 (historically tilled
successional). Treatment 7 plots had also been under cultivation before 1989,
but since that year have been undisturbed, unfertilized, and left to revegetate
with native successional flora. Treatment 7 plots contain 5x5 m micropIots
which have been fertilized every year. These three main treatments plus the
micropIots are described in the following chapters as cultivated, never-tilled,
unfertilized successional, and fertilized successional treatments, respectively.

The most common method for studying AAO populations in soils has
been to estimate numbers of MOS per gram of soil with Most-ProbabIe-Number
(MPN) analysis (Schmidt and Belser, 1982). MPN media tend to select for
nitrifiers that grow fastest under the conditions provided, however, and recovery
efficiency is poor (Belser and Mays, 1982). MPN analysis is uninformative
about in sifu population structures of M03 in soils. Although fluorescent
antibody analysis has been used to study AAO population diversity in soils
(Belser and Schmidt, 1978), fluorescent antibodies can only be prepared for
I AAO strains that have already been cultured, so that uncultured strains are not
detected. In one study, antibodies developed against all isolates from one soil
did not cross-react with half of the isolates obtained from a different soil. AAO
diversity thus appears to be too great for fluorescent antibody analysis to
provide reliable characterizations of population structures in soils from diverse

locations.

4

Nucleic acid-based methods offer significant opportunities for the study of
in sifu populations of AAOs. These methods, which are based on the study of
bacterial DNA rather than the bacteria themselves, do not require the organisms
to be cultured. The first steps in nucleic acid analysis of bacterial communities
are to extract community DNA from soils by lysing the cells and purifying the
DNA that is released. Appendix A at the end of this dissertation describes the
development and evaluation of the DNA extraction and purification methods
Used in this study. AAO populations (typically 104 to 105 per gram) represent
less than 0.01% of total bacterial numbers in soils, which typically range from 1-
5 x 109 per gram (Schmidt and Belser, 1982). Thus, the total amount of
bacterial community DNA from soil will contain only a small fraction of ammonia
oxidizer DNA. Despite proportionally small populations of AAOs in soils,
specific genes or gene fragments from their DNA can be copied in the
polymerase chain reaction (PCR) with oligonucleotide primers that are
complementary to their genes (Schmidt et al., 1991). Gene fragments in the
PCR product mixtures can be separated by inserting individual fragments into
clones of Escherichia coli, these can then be analyzed for their DNA sequences
or for cross-reactivity to other nucleic acid probes (Schmidt et al., 1991). Entire
PCR mixtures can also be analyzed by denaturing gradient gel electrophoresis
(DGGE), which is now considered to be more efficient and informative than
previous cloning and sequencing methods (Muyzer et al., 1993). DNA
sequence variation in these gene fragments should reflect variation in the
population which gave rise to them, assuming there is negligible bias in the
molecular techniques used (Suzuki and Giovannoni, 1996).

Two types of genes could be useful in analyzing variation within AAO
populations. The first type are functional genes that code for enzymes in the

ammonia oxidation pathway. DNA sequences for the genes for ammonia

5
monoxygenase and hydroxylamine oxidoreductase, the two principal enzymes
in this pathway, were made available by McTavish et al. (1993) and Sayavedra-
Soto et al. (1994), respectively. These sequences were obtained from cultures
of Nitrosomonas europaea, which is the most extensively studied AAO strain.
Much of what is known about N. europaea has been generalized as being
representative of all AAOs in the environment, although this may not be a valid
generalization. Two distinct groups of terrestrial AAOs are now recognized on
the basis of phylogenetic variation: 1) the Nitrosomonas group, which includes
all strains of Nitrosomonas and some strains of Nitrosococcus, and 2) the
Nitrosospira group, which includes strains of Nifrosospira, Nitrosolobus, and
Nitrosovibrio. All terrestrial AAOs identified to date belong to the 8-subgroup of
Proteobacteria (Woese et al., 1984), although some marine AAOs
(Nifrosococcus spp.) belong to the gamma subgroup.

The second type of gene that would be useful in characterizing
population diversity of MOS are phylogenetic genes coding for ribosomal RNA
(Figure 1.1). Bacterial rRNA genes are 1500 bases long and contain signature
regions that are specific for various bacterial groups. These regions offer
specific binding sites for oligonucleotide primers or probes for group-specific
PCR. The DNA sequences for the small subunit (16S) rRNA genes of 13 AAO
strains are available in the Ribosomal Database Project (RDP, Maidak et al.,
1994). McCaig et al. (1994), in the laboratory of Jim Prosser at the University of
Aberdeen, Scotland, used the RDP sequences to develop primers specific for
M03 and their close relatives in the B—Proteobacteria. These primers were
used in PCR reactions on community DNA from a wide variety of environmental
samples. The PCR products were used to generate a database of 110 partial
16S rRNA sequences. The region for which these partial sequences were

obtained was 318 bases long and is indicated in Figure 1.1. This region

6

Secondary Structure: small subunit ribosomal RNA

 

  

    

u“ ‘G
‘U U
°’§
no Ac 9’E“°°
. ' . J- _I 1
CU ‘0 l - ‘cc, U U' 0‘3. GC‘A ' :0
‘A‘bcc AucucoAcu“u ‘5 Cbc G U ACGAoc CCUUA uccuuuGUUGCC CGGUC
0 -Il I'IIIII- 0| Oll G IIIIII Il-ollllIIII IOIIIII C
0 acre uocAccuu A c c . uocucc r-
GI AUG I AAO (1.? Go 3.33 | ‘ 6660050 ActioAAaltcucA‘oo “‘5‘ch
" A-u c O - its
-o-c \ A c
-A o u-A
2:8 . cu 9-g E a;
G-C A“ L— G C
G‘U- U U U U ‘ ’c
U A G-C "
6‘ o G A c t -u-A‘ 0" I CI:
A 0 G U I
6_C 9-8 0 G G uc‘flzoo A ‘
'C—G T_.. C-G G-CU -‘AU
C‘U c-o
U-A A 6'” AU‘A
- A o o A
9.0 A u I I... C .
9:; '4 A 6‘ ticrAGAAIIGU : 3.3,
;_U730 ‘C-6 IiI-II- [CG] 10303-0
. 550 U6 G 59 'U G‘c9UUUGC‘GG’I UGA Ug:§
‘ A‘c tua \ u ‘ 8:2 c‘fl 6‘ 0-5-
A ccwscc ocCAucqu cooccAAdc d-A I bc’l C I Uu
-c II-III -IIIIIIIII -III-I,Cc. o—c \cUu ‘°°° u’ A
Ucoooccc GUGUAcAcu 0A uuouuuoc I, uCA ‘E‘Grc c c ‘ uc—cC
UA‘ 600 06 c. A A (Po (3 AG c-c
' A l/UGCA “ U6 - co , o-c
-A
A _ c
6c
c c
UA C O
A C. A I‘Ccu \
‘ C G‘\‘ UU'AU 0 \\ 6‘
UIJ IUU \GC\ LIA uO-C C at.
G I U c \ U' ‘ - ‘ A‘G \
A I 1 GC 0 c use \ G c
I“’o‘l ‘ A U-A I AG c A.
U ’IUrNV- I ‘CGU m S-g- c“‘°\\\c 60
‘q’ a.) G U ' A—g A c” C?
. A U A C" U A
'. cGu o-c G o—c ‘ .
“t c-o nu . CGC\\UGAA‘U
Ag G c-G CA uuGOA 0‘3 V ‘Cuugc' \‘CG I
“IA U1,— -”\t.l U‘Ag:e- a a ’ “‘ [11 Cu \..JG
.r'qi.’ [‘G. ‘UG ' \ G .\ ,,2
.~ .. o I .~
"by ‘6 C05 -0 6c I 1,,”uquccGGGCCCGCACAA°\\‘Cu/‘ 3
_‘AAA5m2:8 ‘0‘ _UcooUAcu iII-~IIIIII \CG " A
. "-Ux, a -0 ”futons“ t-U ‘ Ac‘ucu eccoo c U‘ 06 UG-‘m
“- ‘11' ‘ ‘5 t3u \ I)"J 6‘0” U“ c c 6 cc
GGGUU ”GUI; ‘5C\\\‘A6‘ A‘ C U U U6‘” (3
C II' °III A‘sc .10 C ‘U A A G
. I 3 w. . re 2. .-..
. s \ — u
I AG‘A?‘N CCOCO O I: c .A O c‘ u" c ,u"%
€Uf\‘oe \ . C t: O UAA‘3 ”GOG CZG'
l C‘G « C I (3 C A
‘Aoccua‘ucc‘6c\\6Ac I 5_CIIDUAACCGUAGGGG Ac”o‘u U26
CIII- till a 5, U U """HIA C $3-0
GCGGGU ACCUA C ‘C-G‘ GUUOGCGUCCA GU G O’CU
r A‘CI U.” c c-o‘ 2' “ c G
6‘ x” ‘3-3- U ”c AC
G - c A
‘ c-o A I
TGZC A o o c
U A {106 C
.. izst .
A o no c-o 6°". 8
u x G. cAG—c A d U-
c l A G
c I G A GOD I:
e ’Uc’ “ ‘3’3' I
A GA \AA t '2' r
‘ GUCCAG cuc Ac°oUCAooAA0AAoc 8.3 3
Joe? ‘HII, I-III lllIIoIIIIIIB u—A
‘A GUC‘ c“Sui: coo UG‘éZA ouc uuucuuco U ~ 6
0 \ G I O-U
‘ A A A O I C-O
a” U C C
U I. C‘ I A \ O 1m A-U
cue “cheese C‘ A-U
0U I, GU G ' 2.-U.
Cr A C
cu U ‘A c
cC \ A. q-c
’3‘ ‘ (3U '6U \ 0-;
A c. A u
. ‘ G-C aou
0c UU A—u 0“:
0° \ A ‘3‘0 ‘u-C
\ 5 C-rG A
. . 0-C-
‘ ‘ c’o c-o
U ‘ tIII-ado ‘3 C‘ ‘ .g
\ C \ ”G A- U
C‘\'{ 3-2 ‘0.“
Ocuo\ou U G c-“
O \ O P c C 0
c o-c use—u O
‘ 0‘ G'U UC o o o
. .. 5:2 Escherichia coli
A-u- 1
2-3 I."
\ "' A ~ -
Goo ccucuu“ ‘o°oo°‘ ‘CU‘CUGG - Domain: 3m
-c IIII-II- IeIIo Gilli-I: .
”u cc ooovo coca,| COAUGOCA KI ; PU Mlle
Ad I A ‘ uuu
are A ’ I -
% ~55. (Dukx. gunune
i-if
f-g * February 15, 1993 v2.0
‘C (J01w5)
FIGURE 1.1. Structure of 168 rRNA molecule of Escherichia coli. (From

 

the Ribosomal Database Project (Maidak et al., 1994).

7
corresponds to positions 382 to 500 of the E. coli rRNA molecule.) This

database represented a much greater breadth of gene sequence diversity
among AAOs than previously available in the RDP. The database enabled
Stephen at al. (1996a) to identify four distinct clusters of Nitrosospira and two
clusters of Nitrosomonas. The greater number of rRNA gene sequences thus
made it possible to obtain a more detailed characterization of genetic diversity
among M05 in the B-Proteobacteria.

My dissertation project represents one of the first applications of nucleic
acid-based techniques to the study of M03 in soils. This dissertation was
undertaken as part of two larger community analysis research efforts. The first
effort was that of the Community Diversity Thrust Group in the Center for
Microbial Ecology (CME), a National Science Foundation (NSF) Science and
Technology Center. The research goals of this thrust group are to determine
temporal and spatial patterns of microbial diversity, identify environmental
influences on community structure, and demonstrate effects of community
structure on ecosystem function (CME, 1993). The second research effort, also
funded by NSF, was the LTER project at Kellogg Biological Station, which
integrates the study of above- and below-ground biotic diversity in agricultural
and native ecosystems.

My approach to studying microbial soil community structure was to focus
on AAOs. because of their unique functional role in nitrogen cycling and nutrient
availability. By examining diversity within one bacterial group and relating it to
soil biochemical measurements, I hoped to generate meaningful insights into
the relationships between soil microbial diversity and ecosystem function. I
believed that this approach could circumvent some of the technical and
interpretive problems involved in the study of whole soil microbial communities,

which are characterized by vast diversity and complexity.

8

A grant from the U. S. Department of Education for environmental
microbiology education enabled me to work for five months in Dr. Prosser's
laboratory. During that time, I learned the cloning, sequencing, and probing
techniques used in this laboratory (Embley, 1991). In Aberdeen I also had
access to the partial 16S rDNA sequence database and identified eight
oligonucleotide probe sequences for different AAO subgroups (Stephen et al.,
1996b). After I returned to the U. S., John Stephen and George Kowalchuk
performed DGGE analysis on DNA samples from LTER soils at the Netherlands
Institute of Ecology in Heteren.

My dissertation research involved three phases, each of which
corresponds to a chapter in this dissertation: 1) culturing known M05 and
developing more reliable cultural methods for AAOs from LTER soils; 2)
evaluating nucleic acid probes for functional genes by using genomic DNA from
pure AAO cultures; and 3) developing rRNA gene probes for AAO subgroups
and analyzing diversity of rDNA sequences retrieved from bacterial community
DNA. Each chapter presents the rationale, methods, and results for one of
these research phases. The final section of the last chapter contains a
discussion of the main conclusions from this study. The appendix at the end of
this dissertation contains a published anicle which describes methods
development for bacterial DNA extraction from soils. Key findings from the
following chapters and appendix are summarized as follows.

Chapter Two, Cultural Methods for Studying Autotrophic
Ammonia-Oxldlzlng Bacteria, describes the performance of cultural
methods for enumerating and isolating AAOs from LTER soils. Key findings in
this chapter were: 1) use of lower substrate concentrations in most-probable-
number (MPN) media (10 ppm vs. 100 ppm NH4-N) gave higher recovery of
MOS from LTER soils; 2) NH4-N concentrations in LTER soils ranged from 1 to

9

10 ppm; this may explain why more AAOs were recovered at the lower NH4-N
concentration; 3) MPN counts of M05 from never-tilled soils were consistently
one-tenth lower than MPN counts from cultivated soils during a three-year
period; 4) ratios of MPN counts obtained at 10 ppm vs. 2000 ppm NH4-N were
twentylold higher for AAO populations from never-tilled soils than for
populations in cultivated soils; and 5) use of a lower NH4-N concentration in
soil enrichments was an important factor in the isolation of a novel Nitrosospira
strain from never-tilled soil.

Chapter Three, Evaluating Nitrosomonas europaea functlonal
gene probes In hybrldlzatlon tests with NIfrosospIra DNA, is a
feasibility study for using N. europaea functional gene probes to detect
homologous genes from other AAOs in the pool of DNA from the rest of the
microbial community. Key findings in this chapter were: 1) probe hybridization
signals from genomic DNA varied enough among the MOS tested to require
that probe hybridization tests be performed under conditions of low specificity,
which would increase the risk of nonspecific probe binding to DNA from other
bacteria; 2) hybridization signals with genomic DNA using a gene probe for the
hydroxylamine oxidoreductase enzyme were less variable across the MOS
tested; this means that the probe would be more Specific and reliable than other
functional probes for estimating AAO numbers; and 3) gene sequence
variability and differences in gene copy number make functional probes less
reliable for giving absolute estimates of the sizes of mixed AAO populations;
relative estimates are feasible if AAO numbers are sufficiently high to produce
detectable signal.

Chapter Four, Use of 16S rDNA Analysis to Compare
Dlverslty of Autotrophic Ammonia Oxldlzer Populations In Soils,
describes the use of PCR-based 16S rDNA analysis to compare AAO

10
population structures in cultivated, never-tilled, and successional soils at LTER.
Key findings were 1) AAO gene sequences from never-tilled soils were more
diverse than AAO sequences from cultivated soils; 2) sequences of Cluster 3
Nitrosospira, a phylogenetically distinct group identified by Stephen et al.
(1996a), predominated in samples from cultivated and fertilized successional
soils but were not detected in samples from cultivated and fertilized
successional soils; 3) DGGE banding patterns were spatially reproducible over
an area of 1 km2, with replicate plots for the same treatments giving identical
patterns in two successive years.

Chapter Four's final section discusses the study‘s main conclusions,
relates them to CME and LTER research goals, and interprets them in the
contexts of microbial ecology and agronomy. Conclusions are also based on
data for relevant soil properties that have been obtained as part of the LTER
data collection program. The main conclusions of this study are: 1) sizes of
AAO populations in never-tilled soils are significantly smaller than AAO
populations in cultivated soils by a factor of ten; 2) AAO gene sequences in
cultivated soils exhibit less diversity than AAO sequences in never-tilled soils,
with Cluster 3 Nifrosospira sequences predominating in cultivated soils; 3)
predominance of Cluster 3 Nitrosospira sequences appears to be associated
with nitrogen fertilizer application, because these sequences were also
observed in fertilized but not unfertilized successional soils; 4) effects of AAO
population differences on nitrogen cycling in LTER soils may be reflected in the
consistently higher NH4-N levels observed in never-tilled plots; and 5) although
different AAO populations in cultivated and never-tilled soils exhibit similar net
nitrification rates during in-field nitrification tests, differences in population
function may still be exhibited by the lower percentages of NH4—N converted to
NOa-N during the measurement periods.

1 1

Appendlx A, DNA Recovery from Soils of Diverse
Composition, describes the performance of DNA extraction and purification
methods with eight physically and chemically distinct soils, including soils from
the cultivated and never-tilled plots at the K88 LTER site (CK and NK soils,
respectively). Key findings were 1) bacterial cell lysis efficiency in soils using
the high-NaCI/SDS/heat method varied from 26 to 92%, indicating that this
extraction method recovers the majority of bacterial DNA from some, but not all,
soils; 2) crude DNA yields from cultivated and never-tilled soils were within the
expected ranges based on direct counts of bacteria prior to extraction; this
extraction method thus recovered the majorityof bacterial DNA from these soils;
3) coefficients of variation for crude and purified DNA yields from cultivated and
never-tilled soils were high (about 20%), which could affect reproducibility of
molecular analyses based on this method; 4) DNA yields from never-tilled soils
were approximately four times higher than yields from cultivated soils; and 5)
DNA obtained with these purification methods was of high molecular weight (>
23 kDa) and pure enough for PCR amplification.

Future research that would build on the findings from this dissertation
would include additional replications of these molecular analyses at more
frequent time intervals. Additional replications overtime would help verify the
reproducibility and reliability of these techniques. Any molecular analyses of
soils would be greatly enhanced by measuring 15N transformations in the same
samples to observe relationships between gross nitrification and AAO
population structure. Since the field of molecular microbial ecology continues
to advance, additional primers and probes are likely to become available,
particularly for M05 in the gamma-subgroup of Proteobacteria. The LTER
treatments at the K88 site offer important opportunities for continued research

on the microbial ecology of nitrification.

12
REFERENCES

Belser, L. W. and E. L. Mays. 1982. Use of nitrifier activity measurements
to estimate the efficiency of viable nitrifier counts in soils and
sediments. Appl. Environ. Microbiol. 43:945-948.

Belser, L. W. and E. L. Schmidt. 1978. Diversity in the ammonia-oxidizing
nitrifier population of a soil. Appl. Environ. Microbiol. 36:589-593.

Bock, E., H.-P. Koops, and H. Harms. 1989. Nitrifying bacteria. In H.
Schlegel (ed.) Autotrophic bacteria. FEMS Symposium, Series
Analytic, vol. 42, Federation of European Microbiological Societies.

Center for Microbial Ecology, 1993. Research Findings, Center for
Microbial Ecology, East Lansing, Mich.

Clark, F. E. and E. A. Paul. 1970. The microflora of grassland. Adv. Agron.
22:375-435.

Embley, T. M., 1991. The linear PCR reaction: a simple and robust method
for sequencing amplified rRNA genes. Lett. Appl. Microbiol. 13:171-
174.

Head, I. M., W. D. Hiorns, T. M. Embley, A. J. McCarthy, and J. R. Saunders.
1993. The phylogeny of autotrophic ammonia-oxidizing bacteria as
determined by analysis of 16S ribosomal RNA gene sequences. J. Gen.
Microbiol. 139:1 147-1 153.

Killham, K. 1986. Heterotrophic nitrification. In Nitrification, J. l. Prosser (ed.),
Special publications of the Society for General Microbiology, vol. 20, IRL
Press, Oxford, England, pp. 117-126.

Koops, H.-P. and U. C. Moller. 1992. The lithotrophic ammonia-oxidizing
bacteria. In The prokaryotes, vol. 3.

Kuenen, J. G. and L A. Robertson. 1988. Ecology of nitrification and
denitrification. In J. A. Cole and S. Ferguson (eds.), The nitrogen and
sulfur cycles, Cambridge University Press, Cambridge, U. K., pp. 162-
218.

Kurokawa, M., Y. Fukumori, and T. Yamanaka. 1985. An hydroxylamine
cytochrome c reductase occurs in the heterotrophic nitrifier Anhrobacfer
globiformis. Plant Cell Physiol. 26:1439-1442.

Maidak, B. L., N. Larsen, J. McCaughey, R. Overbeek, G. J. Olsen, K. Fogel, J.
Blandy, and C. R. Woese. 1994. The ribosomal database project. Nucl.
Acids Res. 22:3485-3487.

13

McCaig, A. E., T. M. Embley, and J. l. Prosser. 1994. Molecular analysis of
enrichment cultures of marine ammonia oxidisers. FEMS Microbiology
Lett. 120:363-368.

McTavish H., J. A. Fuchs, and A. B. Hooper. 1993. Sequence of the gene
coding for ammonia monooxygenase in Nitrosomonas europaea. J.
Bacteriol. 175:2436-2444.

Muyzer, G., E. C. de Waal, and A. G. Uitterlinden. 1993. Profiling complex
microbial populations by denaturing gradient gel electrophoresis
analysis of polymerase chain reaction-amplified genes coding for 16S
rRNA. Appl. Environ. Microbiol. 59:695-700.

Nommik, H. and K. Vahtras. 1982. Retention and fixation of ammonium and
ammonia in soils. In Nitrogen in agricultural soils, F. J. Stevenson (ed.),
American Society of Agronomy, Madison, Wisc., pp. 123-171.

Robertson, G. P. 1990. Factors regulating nitrification in primary and
secondary succession. Ecology 63:1561-1573.

Robertson, G. P., J. R. Crum, and B. G. Ellis. 1993. The spatial variability of
soil resources following long-term disturbance. Oecologia. 96:451-
456.

Robertson, G. P., H. P. Collins, S. H. Gage, K. L. Gross, S. J. Halstead, R.
H. Hanuood, K. M. Klingensmith, M. J. Klug, and E. A. Paul. 1996. Long-
term research in agricultural ecology: objectives and establishment of a
site in the U. S. Midwest. Agric. Ecosyst. Env., in press.

Sayavedra-Soto, L. A., N. G. Hommes, and D. J. Arp. 1994. Characterization of
the gene encoding hydroxylamine oxidoreductase in Nitrosomonas
europaea. J. Bacteriol. 176:504-510.

Schmidt, E. L. and L. W. Belser. 1982. Nitrifying bacteria. In A. L. Page, R.
H. Miller, and D. R. Keeney (ed.) Methods of soil analysis, Part 2.
Chemical and microbiological properties, American Society of
Agronomy, Madison, Wisc., pp. 1027-1041.

Schmidt, T. M., E. F. DeLong, and N. R. Pace. 1991. Analysis of a marine
picoplankton community by 16S rRNA gene cloning and sequencing.
J. Bacteriol. 173:4371-4378.

Stephen, J. R., A. E. McCaig, 2. Smith, J. l. Prosser, and T. M. Embley.
1996a. Molecular diversity of soil and marine 16S rDNA sequences

related to 8-subgroup ammonia oxidising bacteria, manuscript
submitted to App. Environ. Microbiol.

14

Stephen, J. R., G. K6walchuk, M. A. Bruns, J. I. Prosser, T. M. Embley, and

J. Woldendorp. 1996b. Analysis of 8-subgroup chemolithotrophic
ammonia oxidiser populations by DGGE, membrane transfer and
hierarchical 16$ rDNA oligonucleotide probing, manuscript in
preparation.

Suzuki, M. T. and S. J. Giovannoni. 1996. Bias caused by template
annealing in the amplification of mixtures of 16S rRNA genes by PCR.
Appl. Environ. Microbiol. 62:625-630.

Woese. C. R., W. G. Weisburg, B. J. Paster, C. M. Hahn, R.S. Tanner, N. R.
Krieg, H.-P. Koops, H. Harms, and E. Stackebrandt. 1984. The
phylogeny of purple bacteria: the beta subdivision. System. Appl.
Microbiol. 5:327-336.

Young, J. L. and R. W. Aldag. 1982. Inorganic forms of nitrogen in soil. In
Nitrogen in agricultural soils, F. J. Stevenson (ed.), American Soc. of
Agronomy, Madison, Wisc., pp. 43-66.

Zhou, J., M. A. Bruns, and J. M. Tiedje. 1996. DNA recovery from soils of
diverse composition. Appl. Env. Microbiol. 62:316-322.

Chapter Two
CULTURAL METHODS FOR AMMONIA-OXIDIZING BACTERIA

INTRODUCTION

Nitrification is the microbial oxidation of ammonia to nitrate, and it leads
to significant nitrogen (N) losses from agricultural systems when nitrate-N is
leached or denitrified (Keeney, 1986). Autotrophic ammonia-oxidizing bacteria
(AAOs) carry out the rate-determining step of nitrification in most soils, but little
is known about AAO population ecology. In this study, AAO population sizes
and structures were evaluated in cultivated and uncultivated soils from the
National Science Foundation Long-Term Ecological Research (LTER) site at
Kellogg Biological Station, near Kalamazoo, Michigan. The LTER site was
established in 1988 to study ecological interactions affecting productivity,
nutrient availability, and biotic diversity in cropped, successional, and native
ecosystems. representative of the Upper Midwest United States (Robertson et
al., 1996). M05 were selected for study in these soils, because they are a key
functional group affecting nitrogen cycling in microbial and plant communities,
and because nucleic-acid-based methods have recently become available for
their analysis (McCaig et al., 1993).

Standard cultural methods for isolating pure cultures of MOS are difficult
and time-consuming due to the long generation times of these bacteria and
their tendency to be overgrown by heterotrophs (Watson et al., 1989). As a
consequence, few pure AAO cultures are available for study, and these cultures

probably do not represent the breadth of AAO diversity in nature. Despite their

15

16
limitations, laboratory culture methods offer an important entry into the study of
AAO populations. The standard cultural method for estimating AAO population
size in soils is the most-probabIe-number (MPN) procedure, in which soil
dilutions are added to replicate tubes containing growth medium composed of
mineral salts and 0.5 g of (NH4)2$O4 per liter (Schmidt and Belser 1982).
Tubes are examined for AAO growth after 4 to 8 weeks incubation (or longer),
and MPN counts are determined on the basis of the highest dilution at which
growth is observed. Some workers have observed significant differences in
MPN counts when using higher or lower concentrations of (NH4)2S04 in
growth media (Donaldson and Henderson, 1989; Suwa et al., 1994). Our study
employed MPN analyses at three different (NH4)2SO4 concentrations to see if
such differences could be observed, and whether these differences could be
related to soil treatment.

In this study, both cultural and nucleic acid-based approaches were used
to evaluate AAO population sizes and structures in soils under different
management regimes. This paper describes the performance of cultural
methods for enumerating and isolating AAOs from LTER soils and for
maintaining and storing AAO stock cultures during development of molecular
techniques for AAO population analysis. Results of nucleic acid-based
analyses of AAO populations in these soils will be reported in a subsequent

paper

MATERIALS AND METHODS
Stralns. Six strains of ammonia-oxidizing bacteria previously isolated
from soils were used in this study. Strains in this collection included three of the
five genera known to occur in soils-Nitrosomonas, Nitrosolobus, and

Nitrosospira. Nitrosomonas europaea ATCC 25978 and Nitrosolobus

17

multifarmis ATCC 25196 were obtained from the American Type Culture
Collection (ATCC). N. europaea ATCC 19718, Nitrosolobus strain 24-C,
Nifrosospira strain NpAV, and Nitrosospira strain Np39-19 were obtained from
Dr. Ed Schmidt of the University of Minnesota. Cultures representing the other
two generauNitrosovibrio and Nitrosococcusuwere not available for this study.

quuld culture medla. A preliminary study was carried out to select a
liquid medium suitable for culture maintenance. Growth of AAO strains was
evaluated over a 2-month period in two types of liquid media (ATCC Medium
929 and ATCC Medium 1573). Media were filter-sterilized after pH adjustment,
because they contained Mg” and Ca“ ions, which could become unavailable
for bacterial growth as a result of complexation during autoclaving (Table 2.1).

Each culture was transferred in triplicate to 16x125 mm screw-capped
tubes containing 5 ml of the two different media and incubated at 25°C without
shaking. Cultures were checked daily for acid production resulting from
ammonia oxidation, which was indicated by color change of the phenol red
indicator from pink to yellow. A sterile solution of 0.05 M K2003 was added
dropwise to cultures that had turned yellow to bring the pH back to 7.5. Cultures
were evaluated for growth consistency among the triplicate tubes and the length
of time each culture remained viable after repeated pH adjustment. Transfers to
fresh medium were made three times during the 2-month period. Cultures of N.
multiformis ATCC 25196 and Nitrosolobus strain 24-C appeared to decline
more rapidly in vigor than the other cultures, as indicated by their tendency to
stop producing acid after two weeks. Since growth of Nitrosolobus strains was
more consistent in Medium 929 than in Medium 1573, the former was chosen

for culture maintenance.

18

TABLE 2.1. Compositions of modified ATCC Medium 929 and ATCC Medium
1573.

 

 

Components Medium 929 Medium 1573
(Nitrosolobus Medium) (Nitrosomonas europaea
per liter distilled water Medium)

per liter distilled water

(NH4)2804 1.32 g1 1.70 g1

MgSO4 380 mg 200 mg

0aCl2.2H20 20 mg 200 mg

K2HPO4 » 87 mg 15 mg

Fe-EDDHA 1 mg 1 m9

Phenol red (0.5% 0.25 ml 0.25 ml
solution)

Trace elements solution 1.0 ml of following mix: 1.0 ml of following mix:
Na2MoO4-2H2O 10 mg 10 mg
MnCl2-4H20 20 mg 20 mg
ZnSO4-7H20 10 mg 10 mg
CoCI2-6H20 0-2 mg 0.2 m9
CuSO4-5H20 --- 2 mg
Distilled water 100 ml 100 ml

Adjust pH to 7.5 with 0.5 M K2003. Filter sterilize with 0.2-u filter.

 

I Corresponds to 20 mM NH4-N (270 ppm) and 26 mM NH4-N (350 ppm) for
Medium 929 and Medium 1573, respectively.

19

All transfers were made in a laminar flow hood; extreme care was taken
to prevent contamination. All lots of liquid medium were checked for
contamination before inoculation by spotting 0.1 ml onto the surface of Plate
Count Agar and R2A Agar (Difco, Detroit, MI). The culture liquid was allowed to
be absorbed by the agar before plates were inverted for incubation at 3200 for
up to one month. Stock AAO cultures were checked every month for
heterotrophic contamination by this method.

Solid medlum. Growth of pure AAO cultures was tested on a solid
medium containing Medium 929 without phenol red and an agar base of 1.5%
Noble Agar (Difco, Detroit, MI) and 1% 0a003 powder. To prepare one liter of
medium, 500 ml double-strength agar base and 500 ml double-strength
Medium 929 were prepared separately. The agar base was autoclaved in a 2-
liter flask containing a heavy stir bar for mixing. When the agar base had been
cooled to about 65°C, filter-sterilized Medium 929 (double-strength) was
carefully poured into the molten agar while the mixture was being stirred on a
hot plate. Approximately 30-40 ml of agar medium were added per petri plate,
since inoculated plates would be incubated for extended periods. Plates were
prepared immediately prior to inoculation. To inoculate a plate, 50 to 100 III of
culture were spotted onto its surface, and the plate was left overnight at room
temperature to allow water from the culture to be absorbed. Plates were
wrapped in parafilm and placed inverted into plastic bags for incubation at
25°C in the dark. The bags also contained water-soaked paper towels, which
were replaced every 2 months, to provide a moist atmosphere for the plates.

Treatments and soils. LTER treatment plots (1 ha each) are
replicated in a randomized block design. Soils from two agricultural treatments
(cultivated Treatment 1 and low-input-cultivated Treatment 3) were used to

provide cultivated soil samples. The cultivated treatment had been a com-

20

soybean rotation from 1989 to 1994, with wheat introduced as a third rotation
crop in 1995. This treatment was conventionally tilled (annual moldboard
plowing, disking, and cultivation), treated with prescribed applications of
herbicides and insecticides, and fertilized with ammonium nitrate. The N
fertilizer was broadcasted at a rate of 84 kg N per he to the corn crop and 56 kg
N per he to the wheat crop. The low-input-cultivated treatment had been a com-
soybean-wheat rotation since 1989, with a leguminous winter cover (34 kg per
ha hairy vetch broadcasted in August) and reduced chemical inputs. This
treatment was given a starter fertilizer application only (ammonium nitrate
alongside of row using 28 kg N per ha). Never-tilled soils were obtained from
Treatment 8 plots that had not been tilled after clearing of the native deciduous
forest in 1959. (Most of the vegetation in these plots now consists of C3
grasses.) Successional soils were obtained from Treatment 7 plots, which were
under a disturbance regime intermediate between the cultivated and never-
tilled treatments. Successional plots had been in cultivated corn and soybeans
for at least 40 years before they were left in 1989 to revegetate with extant flora.
Nitrogen-fertilized micropIots (5m x 5m) within these successional plots were
also sampled to evaluate fertilization vs. disturbance effects. These micropIots
received 125 kg N per ha (ammonium nitrate) broadcasted in July of each year.

Soils at the LTER site are classified as Typic Hapludalfs (U. S. Soil
Classification System, 1992) belonging to the Kalamazoo and Oshtemo soil
series (fine, loamy, mixed, mesic). The A horizons in these soils are typically 20
cm deep. Soils were sampled in April 1992, July 1993, and August 1995 to a
10-cm depth with a 2.5-cm soil corer (approximately 14 g fresh soil per core).
Soil samples in 1992 and 1995 were composites of 20 soil cores per replicate
plot taken near LTER sampling station #1 in each plot. Two replicate plots were

sampled per treatment. Prior to subsampling, soils were mixed inside plastic

21
bags by manually kneading and shaking the bags. Gravel and other debris
were removed from these soils manually. The 1992 and 1995 samples were
thus true replicate samples from two different plots.

Soils in 1993 were subsamples from the large LTER midsummer
sampling done for annual data collection (Robertson et al., 1996). In the larger
LTER sampling procedure, 20 core samples from each of five sampling stations
per replicate plot were composited and mixed by moist sieving (4 mm sieve).
(Cultivated Treatment 1 has six replicate plots; never-tilled Treatment 8 has four
replicate plots.) The mixed samples from all replicate plots were composited
and sieved again prior to removing subsamples for MPN analysis. The 1993
samples were analytical samples from a composited sample from six or four
replicate plots.

All soil samples were stored at 4°C until they were analyzed in the
laboratory. Soil moisture contents were determined from two 10-g subsamples
dried at 110°C for 48 h. All results reported here are based on soil dry weights.

Most-probabIe-number (MPN) analyses. In initial MPN tests, April
1992 soil samples were taken from the low-input-cultivated (in wheat) and
never-tilled treatments (replicate plots 1 and 2). Three different media, based
on the MPN medium described by Schmidt and Belser (1982) were used in
these trials: 1) full-strength Schmidt-Belser (SB) Medium; 2) full-strength 38
Medium with 0.3% bentonite clay (Sigma Chemical 00., St. Louis, MO); and 3)
1/10-strength 88 Medium without clay. The bentonite clay had been added to
one set of MPN tubes to evaluate whether the increased particle surface area
provided by the clay would enhance ammonia oxidizer growth, as had been
previously reported in the literature (Verhagen and Laanbroek, 1991). Tenfold
dilutions of soil were added to 16x125 mm screw-capped tubes containing 5 ml

of medium (five tubes per dilution). Tubes were incubated in the dark for 8

22

weeks at 25°C and examined for color change in the bromthymol blue indicator
from blue to yellow, which indicated a pH of 6.8.

In July 1993, Medium 929 with varying amounts of (NH4)2$04 was used
as the recovery medium for AAOs in soil samples from cultivated (in corn) and
never-tilled plots. These samples were tested with 5-tube MPNs using tenfold
dilutions and media containing 0.05, 0.5, and 5.0 g of (NH4)2$O4 per liter,
corresponding to 1/10X, 1X, and 10X standard NH4-N concentrations,
respectively (Table 2.2). In August 1995, samples were analyzed from
cultivated (in wheat), never-tilled, and successional plots (unfertilized main plot
and fertilized micropIots). For these analyses, 96-well microtiter plates were
used for MPNs (Rowe et al., 1977) by making twofold dilutions (8 wells per
dilution) in 1/10X, 1X, and 20X concentrations of NH4-N.

After subsamples were taken for moisture determination, 10 g moist soil
were blended with 190 ml 100 mM phosphate buffer (pH 7) in a Waring Blendor
for 1 min. Coarse particles were allowed to settle for 1 min, then 10 ml of the
slurry were transferred to a sterile plastic tray for further dilution in the microtiter
plates. Prior to inoculation, all wells of the microtiter plates were filled with 100
pl of the appropriate media. A multichannel pipettor was used to transfer eight
100-ul sample aliquots from the plastic tray to the first row of wells in the
microtiter plates. When sample aliquots had been mixed with the medium in the
first row of wells, 100 pl of this mixture were transferred to the next set of wells,
thus making a 1:2 dilution. A series of 18 twofold dilutions were typically made
after the first 1/20 dilution in the blender.

MPN tubes or plates were incubated in the dark at 25°C for 8 weeks.
Microtiter plates were double-wrapped in parafilm to prevent moisture loss and
packed in plastic bags containing water-soaked pads to maintain a moist

atmosphere. MPN cultures were checked for nitrate as an indicator of ammonia

23

TABLE 2.2. NH4-N concentrations used in MPN recovery media.

 

Medium (NH4)2SO4 NH4-N NH4-N Comments
gper liter mM ppm

 

1/10X 0.05 0.74 10 Concentration which gave
higher MPN recoveries from
forest soils (Donaldson and
Henderson, 1989).

1X 0.5 7.4 100 Standard concentration for
MPNs on soils (Schmidt and
Belser, 1982).

10X 5.0 74 1,000 Concentration which inhibited

NH4-sensitive Nitrosomonas
strains (Suwa et al., 1994).

20X 10.0 148 2,000 Twice as high as the
concentration which inhibited
NH4-sensitive Nitrosomonas
strains (Suwa et al., 1994).

 

24
oxidizer growth. Since the diphenylamine test for nitrate described by Schmidt
and Belser (1982) did not give reliable results, another nitrate test reagent,
Szechrome NB (Polysciences, lnc., Warrington, PA) was used (Jeanette Norton,
personal communication.) Szechrome N8 reagent was prepared according to

the manufacturer’s instructions in a 2:3 mixture of concentrated phosphoric and

sulfuric acids. To check for nitrate, 50 ul of MPN medium were added to the well
of a plate and mixed with 250 III of the Szechrome reagent. After 15 min, the
color of the test mixture was compared to that of nitrate standards ranging from
0.1 ppm to 100 ppm. The reagent formed a blue color in the presence of at
least 0.1 ppm nitrate. Microtiter plates containing uninoculated MPN media
were held for 8 weeks in the same bags as the control plates, and 50-ul aliquots
of these media were used as negative controls.

Enrichments from never-tilled soils. The samples and SB media
used to test initial MPNs on never-tilled soils in 1992 were also used for long-
term enrichment cultures. Replicate 1-liter flasks were inoculated with 1 g moist
soil. Flasks contained 250 ml full-strength SB Medium (with and without 0.3%
bentonite clay) or tenfold-diluted SB Medium containing no clay. After being
incubated in the dark (without shaking) at room temperature for 6 months, the
flask cultures were streaked out onto solid medium. Twelve-week-old MPN
tubes from low-input-cultivated and never-tilled soils were also streaked out
onto solid medium. Inoculated plates were held in moist bags in the dark at
25°C for one to six months and examined for putative AAO colonies. Some of
these colonies were transferred to liquid medium and monitored for subsequent
growth.

Long-term storage methods for AAOs. AAO viability after freezer
storage (~70°0) and room-temperature storage on 929/0a003 plates was

evaluated. One-ml aliquots of 3-week-old cultures in Medium 929 were added

25

to 1 m1 20% sterile glycerol in 17 x 55 mm glass vials (8-ml capacity) and
equilibrated for 15-30 minutes. These mixtures were placed in a dry ice/ethanol
bath, which froze the cultures solid within 2 minutes. Triplicate samples were
tested for outgrowth after the following treatments: (1) equilibration in glycerol;
(2) equilibration, freezing, and one-day storage at -70°C; (3) equilibration,
freezing, and 3-month storage; and (4) equilibration, freezing, and one-year
storage. Frozen glycerol stocks were quick-thawed (within 2 minutes) by
swirling the vials in a beaker of water at room temperature. One-ml of the
glycerol stock was transferred to 5 ml fresh Medium 929 for incubation at 25°C.
Outgrowth was evaluated by noting how many of the tubes turned yellow and
how long the cultures needed to be incubated before they turned yellow. In
some cases, the glycerol stock mixture appeared yellow immediately after
thawing. The pH of these mixtures had to be adjusted to 7.5 before incubation,
otherwise culture viability could not be determined. This color change was
likely artifactual, because uninoculated Medium 929 would occasionally turn
yellow after freezing and thawing.

Growth scraped from one-year-old plates 01929/Ca003 solid medium
was transferred to tubes containing fresh Medium 929. Tubes were checked for
acid production as an indicator of AAO outgrowth.

R E S U LT S
AAO growth on solld medium. Colonies (approximately 0.2 mm in
diameter) could be observed visually after 4 weeks incubation of 929/0a003
plates that had been streaked with stock cultures of N. europaea, N. multifarmis,
Nitrosospira strain Apple Valley, and Nifrosospira strain 39-19. Inoculation of

plates by spotting 50-100 ml aliquots of cultures onto plate surfaces produced

26
lawns of confluent growth after 3 to 4 weeks incubation. Clear zones, indicating
dissolution of 0a003 by acid production, could be observed in the agar below
and surrounding the lawns. Individual AAO colonies were too small to produce
noticeable clearing, although clear zones could be observed under more
heavily inoculated areas after several months incubation.

Individual AAO colonies on these plates were large enough after 4 to 8
weeks incubation to provide sufficient cell material for rep-PCR. This required
at least five colonies, and attempts to use a single colony for PCR were not
successful. Lawns of growth scraped from plate surfaces also yielded sufficient
cell material to allow fixation and embedding for transmission electron
microscopy (Figures 2.1 through 2.4). All cells exhibited the typical
morphological features that have been used to distinguish AAO strains (Watson
et al., 1989). These included intracytoplasmic membranes, carboxysomes, and
typical cell shapes. Plates of 929/0a003 Medium were also inoculated with
enrichment and MPN cultures from LTER soils.

Most-Probable Number (MPN) Analyses. MPN counts from low-
input-cultivated soils using SB Medium were at least tenfold higher than counts
obtained from never-tilled soils for April 1992 samples (Table 2.3). Instead of
enhancing growth, addition of bentonite clay reduced MPN counts by at least
two orders of magnitude for never-tilled soils. The presence of clay did not
appear to significantly affect MPN counts for low-input-cultivated soils. MPN
counts in diluted medium were reduced by factors of 100 and 30-50 for never-
tilled and low-input-cultivated soils, respectively (Table 2.3).

MPN counts of never-tilled soils were also approximately tenfold lower
than counts from cultivated soils in 1994 and 1995 (Tables 2.4 and 2.5). For

soils sampled in 1994, very little or no difference could be seen between counts

27

 

FIGURE 2.1. Transmission electron microscopy photograph of Nitrosomonas
europaea ATCC 25978 showing intracytoplasmic membranes as Iamellae at
the periphery of the cell. (39,000X magnification.)

28

 

FIGURE 2.2. Transmission electron microscopy photograph of Nitrosolobus
multiformis ATCC 25196 showing intracytoplasmic membranes
compartmentalizing cell into a Iobate structure. (29,000X magnification.)

 

FIGURE 2.3. Transmission electron microscopy photograph of Nitrosospira
strain Np AV (Schmidt) showing lack of intracytoplasmic membranes. Cell is
comma-shaped, due to a single spiral turn. (39,000X magnification.)

 

FIGURE 2.4. Transmission electron microscopy photograph of Nitrosospira
strain R1 grown on Noble agar plates containing mineral medium and 0a003,
(29,000X magnification.)

31

TABLE 2.3. Comparative MPN counts (x 103 M03 per gram of soil) obtained in
Schmidt-Belser medium with and without bentonite clay (one soil sample per

replicate plot in 1992).

 

 

Soil 1/10X 1X 1X + Clay

Cultivated low-input (wheat)

Rep 1 2.3 70 1.1

Rep 2 2.3 130 130

Average (n=2) 2.3 100 66
Never-tilled

Rent <0.1* 2.3 <0.1*

Rep 2 < 01* 13 < 0.1

Average (n=2) < 0.1 7.7 < 0.1

 

* no growth in any tube

32

TABLE 2.4. Comparative MPN counts (x 103 M05 per gram dry soil) obtained
in Medium 929 containing 1/10X, 1X, and 10X standard NH4-N concentrations
(two subsamples per composite sample in 1993).

 

Soil 1/10X 1X 10X Ratio of MPN

counts
(1/10X over 10X)

 

Cultivated (corn)

Subsample 1 490 330 79 6.2

Subsample 2 170 79 17 10.0

Average 330 200 48 6.9
Never-tilled

Subsample 1 7.0 1.7 3.3 2.1

Subsample 2 13 22 33 0.4

Average 1 0 12 18 0.6

 

TABLE 2.5. Comparative MPN counts (x 103 M05 per gram of soil) obtained in
Medium 929 containing 1/10X, 1X, and 20X standard NH4-N concentrations
(samples from two replicate plots in 1995).

 

 

Soil 1/10X 1X 20X Ratio of MPN
counts
1/10X over 20X
Cultivated (wheat)
Rep 5 120 100 45 2.7
Rep 6 130 85 80 1.6
Average 1 30 93 63 2.1
Never-tilled
Rep 3 13 33 0.18 72
Rep 4 12 59 0.19 63
Average 1 3 46 0.1 9 68
Unfertilized
successional 74 97 65 1 1.4
Rep 1 46 250 40 1.2
Rep 3
Average 60 170 23 2.6
Unfertilized
Fertilized
successional 190 120 58 3.3
Rep 1
Rep 3 83 410 83 1.0

Average Fertilized 140 270 71 2.0

 

34
in 1X and 10X concentrations of NH4-N. When a higher NH4-N concentration
(20X) was used in 1995, the MPN ratios (1X counts/20X counts) were more than
thirtyfold higher for never-tilled soils than ratios for cultivated and fertilized
successional soils (Table 2.5).

Enrichments. Six-month enrichments of never-tilled soil in flask
cultures were streaked out onto solid medium, and the inoculated plates were
examined after six weeks incubation. Numerous, small (< 0.2 mm), rust-colored
colonies could be observed on plates that had been inoculated with the
enrichments grown in 1/10X SB Medium. Since these colonies were identical
in size and morphology, ten colonies were picked to one tube containing
Medium 929. The resulting culture was confirmed to be free from heterotrophic
contamination and was designated “R1” for “rust-colored.” A few similar
colonies could be observed on plates inoculated with the 1/10X + Clay
enrichment, but plates inoculated with 1X and 1X + Clay enrichments were
overgrown with spreading fungal and bacterial colonies.

When MPN tubes from low-input-cultivated and cultivated soils were
streaked out onto solid medium, most of the resulting growth was obviously
heterotrophic. Approximately 60 small, isolated, putative AAO colonies were
picked to tubes containing Medium 929, but none of these produced viable
cultures after transfer. Heterotrophic colonies were large (> 3 mm), mucoid, and
white, yellow, or orange in color.

Long-term storage of AAOs. Equilibration of cells in glycerol (10%
final concentration), freezing and thawing, and frozen storage reduced the
viability of all AAO strains, as indicated by the incubation periods needed for
cultures to turn yellow (Table 2.6). When equivalent numbers of untreated cells

were transferred to fresh medium, all cultures turned yellow within 1 day, except

TABLE 2.6. Evaluation of effect of freezing and frozen storage on viability of -
AAOs. Numbers in columns are days required for cultures to turn yellow after
thawing and transfer to fresh medium.

 

 

Treatment N. N. Nitrosolobus Nitrosolobus Nitrosospira Nitrosospira
° europaea europaea multifamis strain 24-0 strain Apple strain 39-19
ATCC ATCC ATCC 25196 Valley
25 978 1 9718
None 1 1 1 2-3 1 1
Glycerol 1 2 1-2 3 3 1-2
only
Glycerol + 1 1.28 3-5 38 2-3 2-5a
freezing
One-month 38 3-5 38 3-5 38 3
frozen
storageb
One-year 14-17 14a 45 14-21 17 14-17
frozen
storage

 

aNophlchangeinoneoutofthreettbes.
b Duplicate, rather than triplicate, tubes used.

36
for Nitrosolobus strain 24-0, which grew somewhat more slowly than the other
strains. Exposure of cells to glycerol without freezing retarded cell outgrowth,
because these cultures needed to be incubated 1-2 days longer than untreated
cultures before they turned yellow. Freezing without storage also had a growth-
retarding effect for most of the AAO strains, as did extended storage at -70°C.
N. europaea ATCC 25978 appeared to be the most resistant of the six strains to
these treatments, while N. multiformis ATCC 25196 appeared to be the most
sensitive. All of the AAO strains had at least one tube (out of duplicate or
triplicate tubes) containing nonviable culture after freezing (Table 2.6). Thus, if
freezing is used as a storage method for the MOS, two or three glycerol stocks
may need to be thawed so that at least one viable culture will grow out. The
glycerol stocks prepared in this study yielded viable cultures after 2 1/2 years of
frozen storage, although for some strains (e.g., N. multiformis), two or three vials

had to be thawed before a viable culture was obtained.

DISCUSSION

MPN media containing three different concentrations of (NH4)2SO4 were
used to obtain a more reliable comparison of AAO population sizes than would
be obtained with a single NH4-N concentration. MPN counts of cultivated and
uncultivated soils in Medium 929 were generally higher at the 1/10X (10 ppm)
NH4-N concentration than at the 1X (100 ppm) concentration by factors ranging
from 2 to 10 (Tables 2.4 and 2.5). When the highest MPNs among the three
NH4-N concentrations were considered for each sample, MPN counts of never-
tilled soils were consistently tenfold lower than counts of cultivated soils.

Soil NOa—N and NH4-N measurements support the conclusion that in situ
AAO populations in never-tilled soils were one-tenth as large as populations in

cultivated soils. Measurements made throughout the 1993 and 1994 growing

37

seasons showed that never-tilled soils had very low N03-N levels, ranging from
0 to 1.5 ppm, with a mean of 0.4 ppm. Cultivated soils during the same period
(corn in 1993, soybeans in 1994) had NOa-N levels that were generally tenfold
higher, and these ranged from 0.9 to 14.4 ppm, with a mean of 3.9 ppm (Figure
2.5). NH4-N levels in never-tilled soils exhibited a fourfold variation, ranging
from 4.2 to 18 ppm, with a mean level of 10.2 ppm (Figure 2.6). The mean NH4-
N level was lower in cultivated soils (3.0 ppm), and NH4-N varied more widely,
with a sevenfold variation from 1.5 to 10.2 ppm (Figure 2.5). The low N03-N
levels found concomitantly with higher NH4-N levels in the never-tilled soils
indicate that AAO populations in these soils were not as active in producing
NOa-N as the populations in cultivated soils. This conclusion is compatible with
previous studies showing correlations between AAO numbers and nitrifying
activity in systems permitting visual enumeration by fluorescent probes
(Laanbroek et al., 1993).

The reason for lower populations of M05 in the presence of higher NH4-
N concentrations in never-tilled soils is unclear. One explanation for lower AAO
populations and NOa—N levels in never-tilled soils could be that the higher
amounts of organic matter in these soils tied up the NH4-N and made it less
available to AAOs. This explanation, however, is not supported by the N03-N
and NH4-N data from successional soils (Figures 2.5 and 2.6), which also
contained higher amounts of organic C (Bruns et al., manuscript in preparation).
Another explanation is that some factor(s) in the never-tilled soils inhibited the
growth or activity of MOS (Rice and Pancholy, 1972).

The differences in inorganic N dynamics and MPN counts between
cultivated and uncultivated soils may indicate differences in AAO population
structures. The MPN results for LTER soils can be related to the observations

of Suwa et al. (1994), who obtained estimates of the prevalence of NH4-

38

.32 60.35 30o 5.5 "cocoon
Ana—«3.?»

 

a E
5 0‘0 @ A600 5 3°—
89 magmaflwaa an pea—ease. as:
ezeunofiemNSNSaS—eflsafisneaeafiuia 9:898"
. . T e

 

a a
.. \

 

 

 

 

 

 

 

 

/\ a

as
825+
823+
82p+

ma

 

 

 

 

 

 

38:93 so: a: 88 5383 .32 e5 82 to.
5026092 Sap emote: can >8 cot—3.3.8 3365 mSoEa vacuum Sun 9.353 089.3 :35 none

5 «iota «2: .301 383.6 Eo: 3:952:88: z-moz .3 29:0 chosen 30cm 23 Lotm .32. “En
mam. E mace. CC .3833.» ten .33 3.5.55: .25 “53923 c. «.05. z-moz can: .m.~ £50:

IIOS 5 19" N-EON 5"

39

u damp 60.300 3.0 cm: "gm
A332— 3

5 .3... E .58 5 3%. E
fin 2:9...» 32 a5 2:36 82
Sqfincxsnaaaaaa 8a Sagas“ 8 Sn 8a an an" 2: a: 8g 8g

 

 

 

 

 

 

 

 

 

 

3,—2.5. '0'
'22:. +

 

—r¢:z 2. IT

 

 

 

 

 

 

a

Afiougto. no: ma; no.0 caonxowv 63.. can mm? .8 5026038
Sun 3022. new mac 3335:»; 3365 2525 vacuum $3 9553 336:. 52a suao 5

«Bots etc .301 33:96 E0: 3:982:32: z‘vIz 3,. 20:0 935% 30cm 33 .95 .33 new
mma— 5 £3 33.3303 was .85 3.3.35: .CC $393.3 c. £26. zézz can: .m.~ use:

 

”influx-7min

4O
sensitive AAOs in activated sludges by comparing ratios of MPN counts
obtained at low (20 ppm) vs. high (1000 ppm) NH4-N concentrations. For
sludges with higher N loadings (mean of 1.2 9 total Kjeldahl nitrogen (T KN) per
liter per day), ratios of MPN counts at low vs. high NH4-N concentrations ranged
from 0.6 to 5.7. For sludges with lower N loadings (mean of 0.2 g TKN per liter
per day), the MPN ratios ranged from 3.7 to > 3900. Suwa et al. (1994)
observed that isolates from low-N sludges grew in media containing 20 ppm
NH4-N but not in media containing 1000 ppm NH4-N, while isolates from high-
N sludges grew at either NH4-N concentration. Dominant M03 in low-N
sludges thus appeared to be NH4-sensitive, while dominant M05 in high-N
sludges tolerated 1000 ppm NH4-N and higher.

The 1/10X (10 ppm) and 10X (1000 ppm) concentrations of NH4-N used
in our MPN media were comparable to the low and high concentrations used by
Suwa et al. (1994). The ratios of MPN counts at 1/10X over 10X did not exceed
10 for any of the soils (Table 2.4), suggesting that the type of NH4-sensitive
strains described by Suwa and coworkers did not dominate the AAO
populations in these soils. When we used a higher NH4-N concentration (20X
or 2000 ppm) in subsequent MPN experiments, ratios of counts at 1/10X over
20X were much higher for never-tilled soils than for cultivated soils (T able 2.5).
Overall, the AAOs in LTER soils were not as sensitive to NH4-N as the AAOs
from low-N sludges in the study by Suwa et al. (1994). The differences in MPN
ratios detected with the 20X concentration, however, suggested that AAO
populations in never-tilled soils were more sensitive to NH4-N than AAOs from
cultivated soils. These results could help explain why NH4-N levels remained
consistently lower in cultivated than in never-tilled soils. Cultivated soils may
have selected forAAO populations that tolerated higher and more variable

concentrations of NH4—N, resulting from repeated N-fertilization and release of

41
N-rich root exudates upon crop harvests. This explanation is supported by the
higher variance in inorganic N levels observed for cultivated soils, relative to
never-tilled and successional soils (T able 2.7).

Differences in the 1/10X over 20X ratios between cultivated and
uncultivated soils also appeared to be reflected in the ratios for fertilized and
unfertilized successional soils. Like the cultivated soils, fertilized successional
soils exhibited low ratios. One of the two unfertilized successional soils
exhibited a higher ratio, although it was not as high as the ratios for uncultivated
soils (Table 2.5). This result could be an indication in this successional soil of
an increase in AAOs that were more sensitive to NH4-N. The highest MPN
counts in the unfertilized successional soils, however, had been observed at 1X
NH4-N concentration, rather than 1/10 X concentration. When MPN ratios were
calculated for these soils from the 1X over 20X NH4-N counts, the ratios were
slightly higher than analagous ratios for the cultivated and fertilized
successional soils. These results also suggest that M05 in cultivated and
fertilized successional soils were more tolerant of higher NH4-N concentrations.

The commonly used NH4-N concentration for MPN counting of M05 in
soils has been 7.6 mM, or 100 ppm (Schmidt and Belser, 1982). Of the different
NH4-N concentrations used in our study, the 1/10 X concentration (10 ppm) was
the most similar to measured soil inorganic N levels (NH4-N plus NOa-N)
during the 1993 and 1994 growing seasons. Mean inorganic N levels were 6.9,
10.6, and 3.9 ppm for cultivated, never-tilled, and unfertilized successional soils,
respectively (Table 2.7). Higher recoveries of MOS from cultivated and never-
tilled soils at the 1/10X NH4-N concentration can probably be attributed to the
fact that this concentration was most similar to soil inorganic N levels. MPN
counts for successional soils, on the other hand, were somewhat higher at 100

ppm than at 10 ppm NH4-N in the recovery media (T able 2.5).

42

TABLE 2.7. Inorganic N measurements (N03-N plus NH4-N) during 1993 and

1994 growing seasons in cultivated, never-tilled, and unfertilized successional
soils, with n=30. '

 

 

Soil treatment Minimum Maximum Mean Variance Standard
deviation
Cultivated 2.6 24.6 6.9 17.2 4.1
Never-tilled 4.5 18.3 10.6 1 1.1 3.3
Unfertilized 2.2 6.1 3.9 1.2 1.1

successional

 

43

The use of the lower NH4-N concentration in enrichment media made it
possible to obtain the novel AAO isolate (Figure 2.4) from uncultivated soil,
because this strain appeared to be much more numerous in the enrichments
containing 10 ppm NH4-N than in enrichments with 100 ppm NH4-N
concentrations. It enrichments had been carried out only at the standard NH4-N
concentration, it is unlikely that this AAO would have been isolated. The
enrichment containing 10 ppm NH4-N yielded numerous identical colonies on
solid medium that could be picked to one tube to provide enough biomass to
generate a viable culture. Sequence analysis of its 168 ribosomal RNA genes
indicated that it belongs in a novel clade of Nitrosospira that is distinct from
previously cultured Nitrosospira spp. (Bruns et al., 1995). Solid media have
been used by other workers to isolate AAOs. When Koops and coworkers
employed their solid medium, they were able to isolate eight new Nitrosomonas
spp. (Koops et al., 1991). These successes in obtaining novel AAOs by cultural
methods demonstrate the need to continue and expand the use of cultural
methods in learning more about AAO diversity.

The solid medium used in this study supported the growth of all pure
AAO cultures and could be used to store cultures for several months, as long as
the plates were kept moist. Storage on solid medium offers an alternative to
frozen storage of MOS, which tend to become less viable the longer they are
kept at -70°C (T able 2.6). This medium could also be used to generate larger
amounts of biomass (i.e., from lawns of surface growth) for microscopy or DNA
extraction. Although the solid medium was useful in maintaining AAO cultures
that were already pure, it did not provide an efficient means to purify individual
AAO strains from mixed cultures. Isolated colonies of putative AAOs from
enrichment cultures could be obtained after 2-3 months incubation, as judged

by clear zones around some colonies. However, most of these colonies did not

44
produce viable cultures after they were transferred to liquid medium, probably
because of the low amount of biomass in the colony. A number of these liquid
cultures were “false positive' for growth, because they turned orange in color
but were still negative for nitrite and nitrate after 2-3 months incubation. Other
workers (Prosser, personal communication) have made similar observations
about AAO growth media and attributed them to chemical changes in the pH
indicator.

Cultural methods become much more powerful in characterizing AAO
populations when they incorporate molecular analysis techniques. Growth from
MPN enrichments, for example, could be amplified using 168 rDNA PCR to
identify AAOs that become predominant at different NH4-N concentrations
during incubation. This approach, however, was not used in the molecular
analysis portion of this study. Instead, PCR was directly performed on bacterial
community DNA extracted from LTER soils. This latter approach provides more
information on in situ AAO populations, rather than on the MOS that grow out
from these populations. Results of direct PCR analysis will be reported in Bruns

et al. (manuscript in preparation).

45

REFERENCES

American Type Culture Collection. 1992, Catalogue of Bacteria and
Bacteriophages, 18th edition, American Type Culture Collection,
Rockville, MD.

Ausubel, F.M., R. Brent, RE. Kingston, D.D. Moore, J.G. Seidman, J.A. Smith,
and K. Struhl. 1990. Current Protocols in Molecular Biology. John Wiley
and Sons, New York, NY.

Both, G.J., S. Gerards, and H.J. Laanbroek. 1990. Enumeration of nitrite-
oxidizing bacteria in grassland soils using a Most Probable Number
technique: effect of nitrite concentration and sampling procedure. FEMS
Microbiology Ecology 74:277-286.

Bruns, M.A., J. Stephen, A. McCaig, Z. Smith, and EA. Paul. 1995. Ammonia
oxidizer population structures in cultivated and uncultivated Michigan
soils. Poster presentation at International Society for Microbial Ecology,
August 25-29, 1995, Santos Brazil.

Bruns, M.A., J. R. Stephen, J. l. Prosser, and E. A. Paul. Use of 168 rDNA
analysis to compare diversity of autotrophic ammonia-oxidizing bacteria
in soils, manuscript in preparation.

de Bruijn, F.J. 1992. Use of repetitive (repetitive extragenic palindromic and
enterobacterial repetitive intergeneric consensus) sequences and the
polymerase chain reaction to fingerprint the genomes of Fihizobium
meliloti isolates and other soil bacteria. Appl. Environ. Microbiol.
58:21 80-21 87.

Donaldson, J.M., and GS. Henderson. 1989. A dilute medium to determine
population size of ammonia oxidizers in forest soils. Soil Sci. Soc. Am. J.
53:1608-1611.

Hecht, R.M., RT. Taggert, and DE. Pettijohn. 1975. Size and DNA content of
purified E. coli nucleoids observed by fluorescence microscopy. Nature
54:60-62.

Koops, H.-P., B. Bottcher, U.C. Moller, A. Pommerening-Roser, and G. Stehr.
1991. Classification of eight new species of ammonia-oxidizing bacteria:
Nitrosomonas communis sp. nov., Nitrosomonas ureae sp. nov.,
Nitrosomonas aestuarii sp. nov., Nitrosomonas marina sp. nov.,
Nitrosomonas nifrosa sp. nov., Nitrosomonas eutropha sp. nov.,
Nitromonas oligotropha sp. nov., and Nitrosomonas halophila sp. nov. J.
Gen. Microbiol. 137: 1689-1699.

46

Koch, A. 1981. Growth measurement. In P. Gerhardt (ed.), Manual of Methods
for General Bacteriology. American Society for Microbiology,
Washington, DC. Pp. 179-207.

Laanbroek, H.J., and J.M. T. Schotman. 1991. Effect of nitrite concentration and
pH on Most Probable Number enumerations of non-growing Nitrobacter
spp. FEMS Microbiology Ecology 85:269-278.

Lewis, RE, and D. Pramer. 1958. Isolation of Nitrosomonas in pure culture. J.
Bacteriol. 76:524-528.

Manz, W., R. Arnann, W. Ludwig, M. Wagner, and K.-H. Schleifer. 1992.
Phylogenetic oligodeoxynucleotide probes for the major subclasses of
Proteobacteria: problems and solutions. System. Appl. Microbiol. 15:593-
600.

Rice, E. L., and S. K. Pancholy. 1972. Inhibition of nitrification in climax
ecosystems. American J. of Botany 59:1033-1040.

Rowe, R., R. Todd, and J. Waide. 1977. Microtechnique for most-probable-
number analysis. Appl. Environ. Microbiol. 33:675-680.

Schmidt, EL, and LW. Belser. 1982. Nitrifying bacteria. In A.L. Page, R.H.
Miller, and DR. Keeney (eds.) Methods of Soil Analysis, Part 2. Chemical
and Microbiological Properties, American Society of Agronomy, Madison,
WI, pp. 1027-1041.

Suwa, Y., Y. lmamura, T. Suzuki, T. Tashiro, and Y. Urushigawa. 1994.
Ammonia-oxidizing bacteria with different sensitivities to (NH4)2SO4 in
activated sludges. Water Research 28:1523—1532.

Verhagen, F.J.M., and H.J. Laanbroek. 1991. Competition for ammonium
between nitrifying and heterotrophic bacteria in dual energy-limited
chemostats. Appl. Environ. Microbiol. 57:3255-3263.

Watson, S.W., E. Bock, H. Harms, H.-P. Koops, and AB. Hooper. 1989. Nitrifying
bacteria. In M”. Staley (ed.) Bergey’s Manual of Systematic Bacteriology,
Vol. 3, Williams and Wilkins, Baltimore, MD, pp. 1808-1834.

Chapter Three
EVALUATION OF NITROSOMONAS EUROPAEA GENE PROBE
HYBRIDIZATIONS WITH NITROSOSPIRA DNA

INTRODUCTION

Autotrophic ammonia-oxidizing bacteria (AAOs) convert ammonia to
nitrite in the first, rate-limiting step of nitrification. This step, which is followed by
conversion of nitrite to nitrate by nitrite-oxidizing bacteria, is a critical link in
nitrogen cycling in diverse ecosystems (Bock et al., 1989). AAOs thus represent
a key functional group of bacteria in microbial communities in soils, sediments,
waste treatment systems, building facades, and fresh, brackish, and marine
waters (Watson et al., 1989; Koops and Moller, 1992). Despite their widespread
distribution, AAOs appear to comprise a phylogenetically narrow group of
eubacteria (T eske et al., 1994). To date, the only phylogenetic subdivision
known to contain AAOs of terrestrial origin is the beta subgroup of
Proteobacteria (Woese et al., 1984), although some marine and brackish water
strains fall within the gamma subgroup (Woese et al., 1985). The apparently
close phylogenetic relationship among terrestrial AAOs is likely to be reflected
in the close similarity of functional gene sequences coding for enzymes
involved in oxidation of ammonia, the sole source of energy for AAOs (Book at
al., 1989).

The ammonia oxidation pathway of MOS has been studied almost
exclusively in Nitrosomonas europaea, a culture that waslpurified from soil
enrichments approximately 40 years ago (Lewis and Pramer, 1958). This

isolate appears to be more amenable to laboratory culture than most other AAO
47

48
strains (Prosser, 1989). A large body of literature exists (Hollocher et al., 1981;
Arciero et al., 1991; Hyman and Arp, 1992; Arciero and Hooper, 1993) on the
biochemical characterization of two enzymes involved in energy generation in
N. europaea-~ammonia monooxygenase (AMO) and hydroxylamine

oxidoreductase (HAO). These enzymes carry out the following reactions:

 

 

AMO
(1) NH3 + 02 + 26' + 2H+ > NHzOH + H20
HAOwiIf10554
ase-acceptor
(2) NHZOH + H20 > HN02 + 49' + 4H+

Gene sequences have recently been published for AMO (McTavish et al.,
1993a) and HAO (Sayavedra-Soto et al., 1994) in N. europaea. These
sequences were obtained from PCR products amplified from N. europaea
genomic DNA and primers based on amino acid sequences of purified N.
europaea AMO and HAO proteins. AMO oxidizes ammonia to hydroxylamine,
and HAO oxidizes hydroxylamine to nitrite. Two genes, amoA and amoB, code
for AMO subunits (McTavish et al., 1993a). The gene amoA codes for the
subunit containing AMO's active site. The gene amoB codes for another
polypeptide involved in AMO activity and is located immediately downstream of
amoA. The gene hao codes for the polypeptide unit of HAO (McTavish et al.,
1993b; Sayavedra-Soto et al., 1994), which is a multimer of this unit (Arciero
and Hooper, 1993). Another protein involved in energy generation in N.
europaea is cytochrome c-554, believed to be the immediate electron acceptor
for HA0 (Arciero et al., 1991). The gene coding for cytochrome c-554, hay, is
located 1200 bp downstream of hao (Sayavedra-Soto et al., 1994).

We obtained Escherichia coli clones containing plasmids with inserts

coding for ammonia oxidation enzymes from the laboratory of Dan Arp (Oregon

49

State University). From these inserts we prepared DNA probes for hybridization
with genomic DNA extracted from pure cultures of four terrestrial AAOs within
the Nitrosospira group. The Nitrosospira group is phylogenetically distinct from
the Nitrosomonas group of AAOs in the beta subdivision of the Proteobacteria
and includes the genera NitrosoIobus, Nifrosospira, and Nitrosovibrio. (Head et
al., 1993; Hioms et al., 1995). Our objective was to evaluate the hybridization
signals generated by these functional gene probes against genomic DNA of
other AAOs that have been isolated from soils. This was a first step in
evaluating whether these probes would be reliable indicators of AAO
population size if they were used in hybridization tests with DNA from mixed
AAO populations in soils. Such probe hybridization tests have the potential to
be more rapid in estimating AAO population sizes in environmental samples
than the Most-Probable-Number (MPN) method, which involves 4 to 8 weeks of
incubation (Schmidt and Belser, 1982; Bruns et al., manuscript in preparation).

Other gene sequences of interest in evaluating diversity among ammonia
oxidizers are those coding for enzymes involved in denitrification. Previous
reports have shown that N. europaea produces soluble nitrite reductase
enzyme (Hooper, 1968; Ritchie and Nicholas, 1974) Poth and Focht (1985)
also demonstrated N2O production from NO2' in N. europaea cultures. The
nitrite reductase enzymes found in M05 may be unique to this group, or they
could exhibit similarity to enzymes found in other denitrifying bacteria. To
explore these possibilities, we also applied gene probes for nitrite-reductase
(Smith and Tiedje, 1992; Ye et al., 1993) and nitrous oxide reductase (Viebrock
and Zumft, 1988) from Pseudomonas strains in hybridization tests against

genomic DNA of AAOs.

50
MATERIALS AND METHODS

DNA probes. Probes developed from N. europaea genes for ammonia
oxidation enzymes and Pseudomonas spp. genes for denitrifying enzymes
were used in hybridization studies (Table 3.1). Escherichia coli cells containing
plasmids with probe inserts were grown in batch culture with appropriate
antibiotics to maintain plasmids. Plasmid DNA was extracted from the cultures
and purified in cesium chloride gradients according to standard methods
(Sambrook et al., 1991). Probe fragments were isolated by agarose gel
electrophoresis after digestion of plasmids with appropriate restriction enzymes.
The Gene-Clean kit (Bio-101, lnc., La Jolla, Califomia) was used to purify probe
DNA from the gel. The purified fragments were heat-denatured and labelled
with [a-32PldCTP (3000 Ci/mM, Dupont New England Nuclear Research
Products, Wilmington, Delaware) by using a random priming kit from Boehringer
Mannheim Biochemicals. Unincorporated nucleotides were removed from the
labelled probes with a spun column approach (Sambrook et al., 1991).
Labelled probe was added to hybridization fluid to obtain a final activity of
approximately 106 cpm/ml. Pseudomonas stutzen’ strain JM300 and
Pseudomonas sp. strain G-179 were used as positive controls for hybridizing
with probes for the Cu-type (Ye et al., 1992) and heme-type (Smith and Tiedje,
1993) dissimilatory nitrite reductases, respectively.

Bacterial strains. Six strains of ammonia-oxidizing bacteria
previously isolated from soils were used in this study (Table 3.2). Strains in this
collection included three of the five genera known to occur in soils--
Nitrosomonas, Nitrosolobus, and Nifrosospira. These were obtained from the
American Type Culture Collection (ATCC) and Dr. Ed Schmidt of the University
of Minnesota. Cultures representing the other two genera--Nitrosovibrio and

Nitrosococcus--were not available for this study.

51

TABLE 3.1. DNA probes used in this study

 

 

Probe Organism Protein encoded by Plasmid Probe size Laboratory
designation (probe source) gene with probe source
sequence and
reference
amoA Nitrosomonas Ammonia mono- 0.70 kb D. Arp
europaea ATCC oxygenase (AMO) Kpnl- (McTavish et
1 9718 acetylene-binding EcoRl al., 1993)
polypeptide
amoB Nitrosomonas Polypeptide 0.67 kb D. Arp
europaea ATCC involved in AMO EcoRl- (McTavish et
19718 activity, gene EcoRl al., 1993)
located immediately
downstream of
email
hao Nitrosomonas Hydroxylamine pNH1 10 0.72 kb D. Arp
europaea ATCC oxido-reductase Kpnl- (Sayavedra-
19718 EcoRl Soto et al.,
1994)
hey Nitrosomonas Cytdchrome c-554, pNH055p1 0.73 kb D. Arp
europaea ATCC believed to accept EcoRl- (Sayan
19718 electrons from HAO Bani-ll Soto et al.,
1994)
TN45 probe Nitrosomonas Auxiliary polypeptide pTN45 4.5 kb T.Tokuyama
europaea ATCC possbly involved in EcoRl- (T okuyama
25978 HAO activity EcoRl et al., 1988)
Cu-dNir Pseudomonas Copper-containing pRTc1.9 1.8 kb J. Tiedje
sp. strain G-179 nitrite reductase Chi-CH (Ye et al.,
1993)
Heme-dNir Pseudomonas Heme-containing szGTH 2.4 kb J. Tiedje
sfufzeri JM300 nitrite reductase EcaRl- (Smith and
Barri-ll Tiedje,
1 992)
nosz Pseudomonas Nitrous oxide 1 2 kb J. Tiedje
sfutzeri ZoBell reductase PslI-Psll (V iebrock
and Zunit,

1988)

 

52

TABLE 3.2. Strains of autotrophic ammonia-oxidizing bacteria used in this study

 

 

Strain Source Morphology Comments
Nitrosomonas europaea American Type Rod Neotype, isolated from soil
ATCC 25978 Culture Collection (location unknown) and puriied

Nitrosomonas europaea Dr. Ed Schmidt,

ATCC 19718 University of
Minnesota ~
N'lrosolobus mullifonnis American Type
ATCC 25196 Culture
Collection
Nitrosolobus strain 24-C Dr. Ed Schmidt,
University of
Minnesota
Niflwospira strain Dr. Ed Schmidt,
Apple Valley University of
Minnesota

Nitrosospira strain 39- Dr. Ed Schmidt,
19 University of
Minnesota

Rod

Lobate

Spiral,
curved rod

with serial dilutions (Lewis and
Pramer, 1 958)

Same strain as ATCC 25978,

separate cukure deposited in
ATCC by Dr. Ed Schmidt

Isolated from agricdtural sol in
Minnesota, cultures tend to lose

vigor quickly.

Isolated from agricultural soi in
Minnesota.

Isolated from agricdtural sol in
Minnesota.

 

53

Cultures of three heterotrophic bacteria were also used to provide DNA
for positive and negative controls. The type stain of Arthrobacter globiformis
(ATCC 8010) was included because it is a heterotrophic nitrifier found in soils
(Verstraete and Alexander, 1972). It was of interest to see if DNA from a
terrestrial, heterotrophic nitrifier (Kurokawa et al., 1985) would cross-hybridize
with gene probes for ammonia oxidation enzymes from an autotrophic
bacterium.

AAO batch culture scale-up, DNA extraction and REP-PCR.
Tube cultures of M03 in ATCC Medium 929 were grown to a cell concentration
of 1-2x107 cells per ml (indicated by slight turbidity) by adjusting the pH of the
cultures with 0.05M K2C03 every other day over a 2-week period. One-ml
aliquots of tube cultures were transferred to 50 ml fresh medium in 250-ml
Erlenmeyer flasks, and the pH of these cultures was repeatedly adjusted with
0.1M or 0.3M K2C03 for 2-3 weeks. Five-ml aliquots from the small flask
cultures were transferred to 500 mL fresh medium in 2-L Erlenmeyer flasks, and
these cultures were grown for another 2-4 weeks. All incubation was at 25°C in
the dark without shaking. Cell counts in flask cultures were determined by
examining wet mounts of cultures in a Petroff-Hauser counting chamber (Koch,
1981) at 1000x magnification.

Cells from the large flask cultures were harvested by centrifugation in
250-mL bottles at 5000 x g for 30 min. Cell pellets from 2 to 4 bottles were
resuspended in Medium 929 without (NH4)2S04 and transferred to 30-ml
Oakridge tubes to concentrate bacterial biomass. The miniprep procedure for
extracting bacterial genomic DNA of Ausubel et al., 1990 (page 2.4.1) was
modified by using tenfold larger volumes of all extraction reagents. This
procedure was used to obtain genomic DNAs of all AAO strains and the two

Pseudomonas strains. Genomic DNA of A. globifonnis, which was more

54
resistant to lysis, was extracted according to the method of Visuvanathan et al.
(1989). REP-PCR (de Bruijn, 1992) was also performed on genomic DNA
samples to compare agarose gel rep patterns among M05 in the culture
collection. This procedure, which is based on PCR amplification of conserved
and repeated DNA elements in eubacteria, generates characteristic fingerprints
for each organism and allows comparison and grouping of identical and related
isolates.

Restrlctlon dlgests, Southern blots, and probe hybrldlzatlons.
Genomic DNA preparations were digested with EcoRI, loaded into lanes on 1%
agarose gels, (3 ug DNA per lane) and subjected to electrophoresis (Sambrook
et al. 1989). Visual examination of the relative brightness of UV-illuminated
DNA in a typical agarose gel (Fig. 3.1) confirmed that all lanes contained
equivalent amounts of DNA, except for lanes containing DNA from Nitrosolobus
sp. strain 24-C. This AAO strain grew more slowly and had lower cell densities
than the other AAOs. making it difficult to obtain comparable amounts of
genomic DNA. The amount of Nitrosolobus sp. strain 24-C DNA available for
loading into one lane was half the amount loaded for other bacterial strains.
Subsequent densitometric measurements of this strain's hybridization bands
were doubled to compensate for the lower amount of target DNA available for
probe binding.

Restriction enzyme digestion, electrophoresis, and Southern blotting of
restricted DNA onto Hybond N+ membranes (Amersham Life Sciences, New
York) were carried out as described by Ausubel et al. (1992). The DNA on the
blots was cross-linked by UV light using a Stratalinker (Stratagene, La Jolla,
California). Blots were prehybridized for 16 h in heat-sealed bags containing a

standard prehybridization solution consisting of 50% formamide, 5X Denhardt's

55

134567891012

' _ fl.
2 i s
:I l e
‘ a
‘ .7 f k
I 1
3 1
g: -.

 

iWWWIWH"'WWP"'W

FIGURE 3.1. Ethidium-bromide-stained agarose gel containing EcoRI-digested
DNA of ammonia oxidizers and control bacteria. (Lane 1), Pseudomonas
stutzen’ JM300 ; (Lane 2), blank; (Lane 3) A. globiformis; (Lane 4),
Lambda/Hindlll DNA size marker, (Lane 5), N. europaea; (Lane 6), N.
multifonm's (Lane 7), Nitrosolobus sp. strain 24-C; (Lane 8), Nitrosospira sp.
strain AV; (Lane 9), Nitrosospira sp. strain 39-19; (Lane 10), Lambda/Hlll DNA
size marker; Lane 11, blank; (Lane 12), Pseudomonas sp. strain G-179. All
lanes contained approximately 3 ug genomic DNA, except for Lane 7
containing DNA from Nitrosolobus sp. strain 24-C, with approximately 1.5 ug.

56
solution, 5X SSPE, and 200 mg per ml salmon sperm DNA (Sambrook et al.,
1989). After the prehybridization solution was removed, blots were immersed in
hybridization solution containing 32P-labelled probes for 24 h at 30°C (low
stringency) or 42°C (moderate stringency). After hybridization, blots were
washed once for 15 min at room temperature with 2X SSC (17.53 g of NaCl and
8.82 g of sodium citrate per liter; pH 7.0)-0.1% sodium dodecyl sulfate (808).

Additional washes and hybridizations were performed at different levels
of stringency: (1) hybridization at 30°C and washing at 30°C in 2X SSC; (2)
hybridization at 30°C and washing at 42°C in 0.18 SSC; (3) hybridization at
42°C and washing at 42°C in 0.1x SSC; and (4) hybridization at 42°C and
washing at 55°C in 0.1X SSC. Either X-omat (Kodak) or Dupont NEN film was
exposed to the probed blots using one Cronex Lightning Plus KE intensifying
screen (DuPont) at -70C for 7-10 h prior to development. To facilitate re-
probing, blots were stripped of radioactivity by washing twice for 15 min in
distilled water that had been heated to boiling and amended with 1% SDS,
followed by rinsing in 0.1x SSC for 15 min at room temperature. Membranes
were kept moist in air-tight containers at 4°C between hybridizations to prevent
drying.

Densitometric measurements from autoradiograms. Digitized
images of autoradiogram films were obtained with the CCD camera of a Gel
Print 2000i (BioPhotonics Corp., Ann Arbor, Mich). Hybridization band
intensities were determined using densitometry measurements of the digitized
images with Spectrum IP Lab image processing software (Signal Analytics
Corp., Vienna, Va.). Band intensities measured for Nitrosolobus and
Nitrosospira strains were compared to those of N. europaea to estimate percent

sequence similarities between the probes and homologous genes.

57

RESULTS

Genomic DNA yields from AAO cultures. Cell counts of 4-week-
old AAO cultures were estimated to be 4 x 107 cells per ml at ODeoo of 0.01. At
this cell concentration the theoretical DNA yield from 1 ml of culture was 360 ng,
assuming average cellular DNA content of 9 x 10'15 g per cell (Hecht et al.,
1975). Typical DNA recoveries from 4-week-old cultures were 100-120 ng DNA
per ml culture. DNA recoveries were lower when older cultures (e.g., 6 weeks)
were used for extraction, presumably due to production of extracellular
polysaccharides that interfered with DNA extraction. Slightly improved DNA
recoveries could be achieved with AAO cultures that had just reached slight
turbidity (0.5-1 x 107 cells per ml).

REP-PCR patterns. REP-PCR patterns were obtained from purified
genomic DNA from the six AAO stock strains (Figure 3.2). The REP-PCR
patterns for N. europaea ATCC 25978 and N. europaea ATCC 19718 appeared
identical, suggesting that the strains were the same (de Bruijn, 1992). This
observation was confirmed by D. Pramer (personal communication), who
indicated that the source of the two separate ATCC accessions had been a
culture purified in his laboratory from a soil enrichment obtained from D.L.
Jensen in Denmark (Lewis and Pramer, 1958).

DNA probe hybridization patterns. Genomic DNA was digested
with EcoRI prior to Southern blotting and hybridization tests. N. europaea DNA
exhibited two hybridization bands with the amo/land amoB probes, three bands
with the hao probe, and one band with the hey probe (Figures 3.3 through 3.6).

Other workers have indicated that N. europaea has two copies of the amo gene

58

12345678

3,054 bp .......... >

1 I636 bp --------- >

  
 

1,018 bp .......... >

IlllliI|lII‘IGIIllIIIIIIQIIIIlill‘ILIIIllillllgllllllllll’llllllwllll'IIllll|lllclllIlllllglllllll‘ulll

0

Figure 3.2. REP-PCR patterns of autotrophic ammonia oxidizer DNA
observed in agarose gels stained with ethidium bromide and illuminated
under UV light. Lane 1, Nitrosomonas europaea ATCC 25978; Lane 2
Nitrosomonas europaea ATCC 19718; Lane 3, Nitrosolobus multiformis
ATCC 25196; Lane 4, Nitrosolobus 24-C (Schmidt); Lane 5, Nitrosospira
NpAV (Schmidt); and Lane 6, Nifrosospira 39-19 (Schmidt).

59

.68 x3 5 come a 828; .023 a 5.3353 23:28 scoeetaaafiz .85 2.2.8 figure):
5 .6mm x to 5 come a 9.23; .oeon a gassing 22:28 6:85.93 32: 2.2.8 Sausage»: E a 8: case .9
Bee <zo Zeasdwa £2 Basses as; sea . e 78 case .6 23882.22 .3 25.: ” >< sea .8 23832 5. 2a.: _ 0.3 case

.8 asagz .8 2a.: ” assesses .2 .a 8a.: u 8.32% .2 .: 2a.: H58.“. 55 389.. so; areas .228 Be segues 3855

2.3285.» .e <zo 855.6 Bea <25 5.; 852...»; <20 2588 a sea 22.58 $2. 9528.91 523833 .3 mason.

 

 

 

 

Evmm; [www.—
INNO.N . INNO.N
‘
,-Omm.m 10mm.m
-Emé . -26;
£33“ 3
’7
-wNN._.N -wNN pN

 

 

 

 

 

60

.68 x8 a. come .a 8.5%. 69¢ .a 5.3.252. 2.3.2.8 6525.23... .28: 32.3

5.3.252. 3. .63 x... a. cow... a 9.28... deem .e 5.3.2.2.... 22.2.8 252526. 5.8.. 23:3 5.3.2.2... a. H3 82 ea.
3. 32a <zo 3_Ba_.a~o 5.; 82.22... 22.. 22m .e Ia 52.» .8 2.88822 .3 2a... .2. case .3 2282.2 2.. 2.3. 0.2 2.2....
.3 3382.2 .6 2a... 2:22.232 .2 .a e5... .8328 .2 .2 Sq... UEeem 5.; 8.8m... 2o; areas. .228 2a 23.3.6 3822..
8.86....» .e <zo 2288 .32.. e2. 5.; 22.2.2. <zo 0.288 .e no... 29.58 22. 3.28.9: 223865.. in $50.".

 

 

 

 

-33.. ' . .33..
El 1.... . .
-Zo.~ I... . -RoJ
I! .635 5,; . $3.6
$3.2 $3;
,
1a.... ’
if '
-mNNJN -325
5%.... . .2.

 

 

 

 

 

emu. 3. mean. a.

61

Gem .2... a. come a 828.... .021... a 3.3.3.2.... 223:8 .8855? 8...... 2.2.8 3.3.3.2...

3. .63 x8 a. come a 9.28; doom 5 8.3.2.2.... 2522.8 3:85.22 .35 2.2.8 53.82.... a. ”3 33.2... .e. 8e...
<zo 8.23.33 5.... 83.25.... as... sea . 2-8 52... .8 38882.2 .3 8e... . >< ...2.a .8 2283.2 6.3 52..» .8 8382.2
.3 9.3. .3 e5... . 2:22.22... .2 .3 83. . 8828 .2 .3 2.3. ”E8... 2...... 8.8a... 23.8.. .228 a... 23.2.6 282.... 028.85..
.e <20 2288 855:8 ”film .32.. .2. 5.; 83.32... <zo 8.68.. .e 2e... 22.58 22. 2228.9... 223.2%... 3.... mass...

 

 

-53.. al.. -23..

3.8.3 -333
- i... '

10mm,.m nemm.m

AS... .23...

33.5 . -33.:

 

 

 

 

 

 

62
3.4.3.. 3,; 5 '7 r 2 a?" 34W. '..-'*’...7..7.. ’13.}? ”.
. g ,. ' L

I Q '1' 1‘ :7 1
‘3'
3. - \
H , 5 ' -
. "~ w» . .
.5” y. 1 ‘1‘" I Q
A «mg»? .5
"1': .- {a
‘

Fiwre 3.6a. Southern blots of AAO genom'c DNA hybridzed with amok probe (left)

and amoB probe (right). l-Mxidzation was carried out at 30°C. Blots were washed at
30°C in 2X SSC buffer. Autoradogam films were exposed for 6.5 h. Sources of
genomic DNA were N. europaea (Late 5); N. nwltiformis (Lane 6); Nitrosolobus sp.

strain 24-C (Lme 7); M'trosospfm sp. strain AV (Late 8); Nitrosospim sp. strain
39-19 (Late 9).

'L-f"~:;‘:‘r‘:=r.¢r crqrgrrvx-
. . I..’.. . ._ ‘._..., _.. _ -- ~-- ..-.: .. ..,, .
ii

.5

Fugue 3.6b. Southern blots of AAO genomic DNA hybridzed with amaA probe (left)
and amoB prd>e (rigit). Hybridzation was carried out at 42°C. Blots were washed at
42°C in 0.1x SSC buffer. Autoradogam films were exposed for 9.5 h. Sources of
genomic DNA wae N. europaea (Lane 5); IV. multifamis (Lane 6); Nitrosolobus sp.

strain 24-C (Lme 7); M'tmsospira sp. strain AV (Lme 8); M'tmsospira sp. strain
39—19 (Lme 9).

63
(McTavish et al., 1993a), three copies of the hao gene (McTavish et al., 1993b),
and three copies of the c-554 gene (Sayavedra-Soto et al., 1994).

The number of hybridization bands and the locations of each band were
identical for the amoA and amoB probes for all AAO strains tested (Figure 3.6).
In N. europaea, amoB is located immediately downstream of amoA (McTavish
et al., 1993). The identical hybridization patterns obtained with amoA and
amoB probes on DNA from all other AAO strains indicates that these genes are
also linked in these strains. Norton et al. (1996) have reported that N.
multiformis and Nitrosospira sp. strain AV each contain three copies of the
amoA gene.

Comparisons of summed band Intensities. To evaluate whether
N. europaea gene probes would provide reliable estimates of population sizes
of other terresterial AAOs. the intensities of all hybridization bands for each
strain, under lowest stringency conditions, were summed to obtain an estimate
of total probe signal from each AAO (Table 3.3). The total probe signal depends
on both gene copy number and percent similarity between the DNA probe and
corresponding genes in different strains. Total probe signal of N. europaea was
designated as 100%, because all probes were derived from N. europaea DNA
sequences. Total probe signals from the other four AAO strains (Table 3.3)
were divided by the total probe signal from N. europaea to obtain percentage
values (Table 3.4). These percentages reflect the relative signal that could be
obtained from one genome of an AAO strain when it is hybridized with the
corresponding N. europaea-derived probe. In general, N. multifonm's exhibited
the lowest percentages (20 to 48%) of the four AAO strains tested. The hao
probe gave the highest overall percentages ( 48 to 68%) among the four strains
(Table 3.4).

TABLE 3.3. Summed intensities1 of all hybridization bands obtained with each probe from
genomic DNA of five ammonia oxidizer strains. Numbers in parentheses indicate the numbers of
hybridization bands contributing to total intensities. Hybridizations were carried out under low

stringency conditions (hybridization at 30°C, washing at 30°C with 2X SSC.)

 

Ammonia oxidizer strait amoA probe amoB probe hao probe hcy probe

 

Nam cum“ 2 220,500 (2) 365,300 (2) 580,600 (3) 277,300 (1)
Nitrosolobus multifor'mis 80,500 (1) 74,400 (1) 276,400 (2) 128,400 (1)
Nitrosolobus sp. strain 24-c 166,600 (2) 46,700 (2) 372,200 (3) 75,400 (1)
Nmmpira sp. strain Av3 158,300 (3) 114,200 (3) 394.400 (3) 75,500 (1 )
Mimeoapira sp. strain 3919 93,500 (1) 107,300 (1) 365,500 (3) 165.300 (3)

 

1 Densitometric measurements of bands in digitized images of autoradiogram films that had been
exposed to Southern blots hybridized with 32-P-Iebeiied probes (Figures 3.3, 3.4, and 3.5).

2 Nitrosomnas europaea contains 2 copies of the amo genes (McTavish at al., 1993a and
1993b), 3 copies of the hao gene, and 3 copies of the cytochrome c-455 (hcy) gene (Sayavedra-
Soto et al., 1994).

3 Nitmsospira sp. strain AV contains 3 nearly identical copies of the amoA gene (Norton et al.,
1 996).

TABLE 3.4. Summed band intensities for N. multiformr's, Nitrosolobus sp. strain 24-C,
Nitrosospira sp. strain AV. and Nitrosospira sp. strain 39-19, expressed as percentages of the
summed intensities for N. europaea (T able 3.3).

 

Ammonia oxidizerstra'n amoAprobe amonrobe hao probe hcyprobe

 

N. europaea 100% 100% 100% 100%
N. mdlifomiis 36 20 48 . 46
Nitrosolobus sp. strain 24-0 76 52 64 27
Nitrosospira sp. strain AV 72 31 68 27

Nitrosospira sp. strain 39-19 45 29 63 59

 

66

No hybridization was observed between any of the N. europaea probes
and DNA from A. globifomvis, P. stutzeri strain JM300, or Pseudomonas sp.
strain G-179 (data not shown).

Hybridization results with other DNA probes. No hybridization
was observed between DNA from the four AAO strains and the TN45 probe
derived from N. europaea (Table 3.1). This probe thus appears to be derived
from a DNA sequence unique to N. europaea among the strains tested. No
hybridization was observed with DNA from any AAO and the probes for the
heme dNir or nitrous oxide reductase. Slight cross-hybridization was observed
between the probe for the Cu-type dNir and DNA from Nitrosolobus sp. strain
24-C and Nitrosospira sp. strain AV under moderate stringency conditions

(Figure 3.7). No hybridization was seen with DNA from the other AAOs.

DIS C U S S i O N

Results from the first comparative hybridization tests between a suite ofN.
europaea functional gene probes and genomic DNA from four representatives
of the Nitrosospira group of ammonia-oxidizing bacteria are presented. These
results provide important information for understanding how molecular analysis
techniques may be applied to the study of heterogeneous AAO populations in
soils and other environmental samples. The key findings in this study were that
DNA from all four Nitrosospira strains hybridized with each N. europaea
functional probe under low-stringency conditions (Figures 3.3, 3.4, 3.5), except
for the TN45 probe. Total hybridization signal at low stringency from each of the
four strains ranged from 20% to 76% of total signal from N. europaea DNA
(Table 3.4). No hybridization was observed for amo and hao probes under
higher-stringency conditions (hybridization and washing at 42°C), except for

very faint bands between DNA from Nitrosospira strain Np 39-19 and probes for

21.226 bp--..> .

5,148 bp.___> :
41973 bp.-..>

2'027 bp—-..> ‘-
11584mm.

 

FIGURE 3.7. Autoradiogram photograph of Southern blot hybridized with nir
probe for a copper-type nitrite reductase gene obtained from Pseudomonas sp.
strain G-179 (Ye et al., 1992). The blot contained EcoRl-digested DNA from the
following cultures: Pseudomonas sp. strain G-179 (Lane 1); Lane 2 blank;
Lambda DNA marker (Lane 3); Nitrosospira sp. strain 39—19 (Lane 4);
Nitrosospira sp. strain AV (Lane 5); Nitrosolobus sp. strain 24-C (Lane 6); N.
multiformis (Lane 7); N. europaea (Lane 8); Lambda DNA marker (Lane 9); A.
globiformis (Lane 10); Lane 11 blank; Pseudomonas stufzeri strain JM300 (Lane
12). The positive control. Pseudomonas sp. strain G-179, exhibited the expected
hybridization band. Note that two AAO strains, Nitrosospira sp. strain AV and
Nitrosolobus sp. strain 24-C, exhibited slight hybridization with the probe.

68
amoA (Figure 3.3) and amoB (Figure 3.6). The hcy probe also produced very
faint bands with DNA from all four strains under higher-stringency conditions
(Figure 3.5).

The results indicate that N. europaea DNA probe hybridization with
microbial community DNA would have to be carried out under low-stringency
conditions to detect other AAOs belonging to the Nitrosospira group. Use of
low-stringency conditions with N. europaea DNA probes would be particularly
critical when analyzing community DNA from soils, because cultural and
molecular studies have shown that Nitrosospira spp. are the predominant AAOs
in many soils (Hiorns, et al., 1996; Stephen at al., 1996; Bruns et al., manuscript
in preparation). In these soils, probes developed from functional genes derived
from Nitrosospira spp. may be more reliable for AAO population studies. it is
likely that probe hybridizations would still have to be carried out at low
stringencies to counteract the variability in functional gene homology (Norton et
al., 1996) observed among AAO strains.

The need for low-stringency hybridizations with functional probes raises
the problem of potential nonspecific binding of the probes to DNA from similar
genes in unrelated organisms. Methane oxidizers, for example, have DNA that
may cross-hybridize with probes for ammonia monooxygenase, because this
enzyme has substrate specificity similar to that of the particulate methane
monooxygenase (McTavish et al., 1993). Since methane-oxidizing bacteria are
also found in soils, DNA from these populations may also bind to amo probes
under low-stringency conditions. For this reason, an hao probe is likely to be
more specific for AAOs than an amoA probe (D. Arp, personal communication).
The hao probe exhibited the highest total hybridization signal of the four N.
europaea probes tested in this study (Table 3.4), suggesting that this gene is
more conserved among AAOs. Of these four probes, the hay probe would be

69
expected to have the most cross-reactivity with DNA from other other bacteria.
because the sequence for the c-554 gene shares homology to genes coding for
other heme cytochromes (Sayavedra-Soto et al., 1994). Under the higher-
stringency conditions used in our study, the hay probe eXhibited greater cross-
reactivity with DNA from all four Nitrosospira strains than the other three probes
(Figure 3.5).

Functional gene probe hybridizations with community DNA cannot be
considered to give absolute estimates of AAO population sizes in mixed
communities. A key problem in using probe hybridizations to estimate AAO
population size stems from our lack of knowledge about the extent of cross-
hybridization between DNA probes that have been developed from cultured
AAOs and the genomic DNA of uncultured AAOs. As a group, the MOS are
extremely difficult to culture and purify, and only 10 to 20 different cultures of
M08 are available in international culture collections. In addition, the
physiological and biochemical characteristics of N. europaea have been
assumed to be generalizable to other AAOs (Prosser, 1989). Physiological
diversity among AAOs may be much greater than previously assumed. The lack
of any cross-hybridization between the TN45 probe sequence from N. europaea
and DNA from the Nitrosospira strains could allude to biochemical attributes in
the Nitrosomonas group that are absent in the Nitrosospira group. Similarly, the
cross-hybridization between Pseudomonas dNir probes and DNA from two out
of four Nitrosospira strains (Figure 3.7) may be an indication of broader
physiological diversity within this group. Continued efforts to modify enrichment
conditions and improve cultural procedures may result in the isolation from soils
of AAO bacteria which belong to phylogenetic divisions other than the beta
subgroup of Proteobacteria. The ammonia oxidation enzymes of such AAOs

would likely have undergone greater divergence and would exhibit less

70
homology to N. europaea DNA probes than the terrestrial strains used in this
study.

Hybridization tests in this study were performed only on genomic DNA
from pure AAO cultures so that cross-reactivity between N. europaea gene
probes and DNA from Nitrosospira representatives could be compared directly.
Since no attempt was made to apply these probes to community DNA extracted
from soils. this study did not address the possibility that AAO numbers in soils
are too low to produce detectable hybridization signals. The lowest detection
limit for high-activity radioactive probes with single-copy targets of 1 kb is 104
organisms per gram of soil (Holben et al., 1988; Harris, 1994). Since MPN
counts of MOS in soils may be 104 or fewer per gram (Schmidt and Belser,
1982), some AAO populations would be too small to be detected with functional
probes. Thus, it would be more feasible to use the probe hybridization
approach with community DNA extracted from soils expected to have high AAO
populations (e.g., soils that have been amended with ammonia-based fertilizers
or animal manures; high-nitrogen-ioaded sludges). Probe hybridizations may
be most useful in situations where comparisons of relative population sizes,
rather than absolute counts, can be used to track spatial, temporal, or treatment

vanafions.

ACKNOWLEDGMENTS
I would like to thank Dr. Dan Arp and Dr. Norman Hommes, of Oregon
State University, for the clones containing probe inserts, and Dr. Ed Schmidt, of

the University of Minnesota, for AAO cultures.

71 .
REFERENCES

Arciero, D. M. and A. B. Hooper. 1993. Hydroxylamine oxidoreductase from
Nitrosomonas europaea is a multimer of an octa-heme subunit. J. Biol.
Chem. 268:14645-14654.

Arciero, D. M., C. Balny, and A. B. Hooper. 1991. Spectroscopic and rapid
kinetic studies of reduction of cytochrome c-554 by hydroxylamine
oxidoreductase from Nitrosomonas europaea. Biochemistry 30:11466-
11472.

Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. A. Smith, J. G.
Sideman, and K. Struhl (ed.) 1987. Current protocols in molecular
biology. John Wiley 8. Sons, Inc., New York, N. Y.

Bock, E., H.-P. Koops, and H. Harms. 1989. Nitrifying bacteria. In Autotrophic
Bacteria, H. G. Schlegel (ed.), FEMS Symposium, Federation of
European Microbiological Societies, Series Analytic, vol. 42, pp. 81-96.

deBruijn, F. J. 1992. Use of repetitive (repetitive extragenic palindromic and
enterobacterial repetitive intergeneric consensuses) sequences and the
polymerase chain reaction to fingerprint the genomes of Rhizobium
melilofi isolates and other soil bacteria. Appl. Env. Microbiol. 58:2180-
2187.

Harris, D. 1994. Analyses of DNA extracted from microbial communities. In
Beyond the biomass: Compositional and functional analysis of soil
microbial communities. K. Ritz, J. Dighton, and K. E. Gilles (eds), British
Society of Soil Science, John Wiley and Sons, New York, NY, pp. 111-
118.

Head, i. M., W. D. Hiorns, T. M. Embley, A. J. McCarthy, and J. R. Saunders.
1993. The phylogeny of autotrophic ammonia-oxidizing bacteria as
determined by analysis of 16S ribosomal RNA gene sequences. J. Gen.
Microbiol. 139:1 147-1 153.

Hecht, R. M., R. T. Taggert, and D. E. Pettijohn. 1975. Size and DNA content of
purified E. coli nucleoids observed by fluorescence microscopy. Nature
54:60-62.

Hiorns, W. D., R. C. Hastings, l. M. Head, A. J. McCarthy, J. R. Saunders. R. W.
Pickup, and G. H. Hall. 1995. Amplification of 168 ribosomal RNA genes
of autotrophic ammonia-oxidising bacteria. J. Gen. Microbiol. 141:2793-
2800.

Holben, W. E., J. K. Jansonn, B. K. Chelm, and J. M. “Radio. 1988. DNA probe
method for the detection of specific microorganisms in the soil bacterial
community. Appl. Env. Microbiol. 54: 703-711.

72

Hollocher, T. C., M. E. Tate, and D. J. D. Nicholas. 1981. Oxidation of ammonia
by Nitrosomonas europaea: definitive 180-tracer evidence that
hydroxylamine formation involves a monooxygenase. J. Biol. Chem.
256:10834-10836.

Hooper, A B., D. M. Arciero, A. A. DiSpirito, J. Fuchs, M. Johnson, F. LaOuier,
G. Mundfrom, and H. McTavish. 1990. Production of nitrite and N20 by
the ammonia-oxidizing nitrifiers. In Nitrogen fixation: achievements and
objectives, Gresshoff, Roth, Stacey, and Newton (eds), Chapman and
Hall, New York, NY.

Hyman, M. R., and D. J. Arp. 1992. 1402H2- and 14COz-labelling studies of the
de novo synthesis of polypeptides by Nitrosomonas europaea during
recovery from acetylene and light inactivation of ammonia
monooxygenase. J. Biol. Chem. 267:1534-1545.

Klotz. M. G., and J. M. Norton. 1995. Sequence of ammonia monooxygenase
subunit a in Nitrosospira sp. NpAV. Gene 163:159—160.

Koch, A. 1981. Growth measurement. in P. Gerhardt (ed.) Manual of methods for
general bacteriology. American Society for Microbiology, Washington,
009 pp- 179'207.

Koops, H.-P. and U. C. Moller. 1989. The lithotrophic ammonia-oxidizing
bacteria. In The prokaryotes, pp. 2625-2637.

Kurokawa, M., Y. Fukumori, and T. Yamanaka. 1985. A hydroxylamine-
cytochrome 0 reductase occurs in the heterotrophic nitifier Arthrobacfer
globiformis. Plant Cell Physiol. 26:1439-1442.

Lewis, R. F. and D. Pramer. 1958. isolation of Nitrosomonas in pure culture. J.
Bacteriol. 76:524-528.

McTavish H., J. A. Fuchs, and A. B. Hooper. 1993. Sequence of the gene coding
for ammonia monooxygenase in Nitrosomonas europaea. J. Bacteriol.
1 75:2436-2444.

Norton. J. M.. J. M. Low, and M. G. Klotz. 1996. The gene encoding ammonia
monooxygenase subunit-A exists in nearly identical copies in
Nifrosospira sp. NpAV. FEMS Microbiology Letters, in press.

Prosser, J. l. 1989. Autotrophic nitrification in bacteria. Adv. Microbiol. Physiol.
30:125-181.

Poth, M., and D. D. Focht. 1985. 15N kinetic analysis of N20 production by
Nitrosomonas europaea: an examination of nitrifier denitrification. Appl.
Environ. Microbiol. 49:1 134-1 141.

73

Ritchie, G. A F., and D. J. D. Nicholas. 1974. The partial characterization of
purified nitrite reductase and hydroxylamine oxidase from Nitrosomonas
europaea. Biochem. J. 138:471-480.

Sambrook, J., E. F. Fritsch, andT. Maniatis. 1989. Moiecularcloning: a
laboratory manual. 2nd ed., Cold Spring Harbor Laboratory, Cold Spring
Harbor, N. Y.

Sayavedra-Soto, L. A., N. G. Hommes, and D. J. Arp. 1994. Characterization of
the gene encoding hydroxylamine oxidoreductase in Nitrosomonas
europaea. J. Bacteriol. 176:504-510.

Schmidt, E. L. and L W. Belser. 1982. Nitrifying bacteria. In A. L. Page, R. H.
Miller, and D. R. Keeney (eds) Methods of soil analysis, part 2. Chemical
and microbiological properties. American Society of Agronomy, Madison,
WI. Pp. 1027-1041.

Smith, G. B. and J. M. Tiedje. 1992. isolation and characterization of a nitrite
reductase gene and its use as a probe for denitrifying bacteria. Appl. Env.
Microbial. 58:376-384.

Stephen, J. R., G. Kowalchuk, M. A. Bruns, J. l. Prossesr, T. M. Embley, and J.

Wollendorp. Analysis of B—subgroup chemolithotrophic ammonia oxidizer
populations by DGGE, membrane transfer, and hierarchical
oligonucleotide probing of 16S sequences. Manuscript in preparation.

Teske, A., E. Aim, J. M. Regan, S. Toze, B. E. Rittmann, and D. A. Stahl. 1994.
Evolutionary relationships among ammonia- and nitrite-oxidizing
bacteria. J. Bacteriol. 176:6623-6630.

Tokuyama, T., Y. Tomita, and R. Takahashi. 1988. Cloning and expression of
the hydroxylamine oxidase gene of Nitrosomonas europaea in
Pseudomonas putida. Hakkokogaku 66:103-107 (in Japanese).

Verstraete, W. and M. Alexander. 1972. Heterotrophic nitrification by
Arthrobacter sp. J. Bacteriol. 1 10:955-961.

Viebrock, A., and W. G. Zumft. 1988. Molecular cloning, heterologous
expression, and primary structure of the structural gene for the copper
enzyme nitrous oxide reductase from denitrifying Pseudomonas sturzeri.
J. Bacteriol. 170:4658—4668.

Visuvanathan, S., M. T. Moss, J. L. Stanford, J. Hermon-Taylor, and J. J.
McFadden. 1989. Simple enzymic method for isolation of DNA from
diverse bacteria. J. Microbiol. Meth. 10:59-64.

74

Watson, S. W., E. Bock, H. Harms, H.-P. Koops, and A. B. Hooper. 1989.
Nitrifying bacteria. In J. T. Staley (ed.) Bergey‘s Manual of Systematic
Bacteriology, Vol. 3, Williams and Wilkins, Baltimore, MD, pp. 1808-1834.

Woese. C. R., W. G. Weisburg, B. J. Paster, C. M. Hahn, R. S. Tanner, N. R.
Krieg, H.-P. Koops, H. Harms, and E. Stackebrandt. 1984. The
phylogeny of the purple bacteria: the beta subdivision. System. Appl.
Microbiol. 5:327-336.

Woese. C. R., W. G. Weisburg, C. M. Hahn, B. J. Pastor, L. B. Zablen, B. J.
Lewis, T. J. Macke, W. Ludwig, and E. Stackebrandt. 1985. The
phylogeny of the purple bacteria: the gamma subdivision. System. Appl.
Microbiol. 6:25-33.

Ye, R. W., M. R. Fries, S. G. Bezborodnikov, B. A. Averill, and J. M. Tiedje. 1993.
Characterization of the structural gene encoding a copper-containing
nitrite reductase and homology of this gene to DNA of other denitrifiers.
Appl. Env. Microbiol. 59:250-254.

Chapter Four
USE OF 168 rDNA ANALYSIS TO COMPARE DIVERSITY OF
AUTOTROPHIC AMMONIA OXIDIZER POPULATIONS IN SOILS

INTRODUCTION

Autotrophic ammonia-oxidizing bacteria (AAOs) are a key functional
group within soil microbial communities, because they carry out the rate-
determining step in nitrification. Terrestrial AAOs represent a narrow
phylogenetic group within the B-Proteobacteria (Head etal., 1993; Teske et al.,
1994), and they are known to be difficult to enumerate and isolate (Schmidt and
Belser, 1982). Their distinctive functional and phylogenetic characteristics
increase the likelihood that oligonucleotide probes and primers developed from
their genomic DNA sequences will be specific and useful for analyzing DNA
from AAO populations in complex communities. Genes coding for small subunit
ribosomal RNA (16S rDNA) have been used most extensively in phylogenetic
analysis of bacteria (Woese et al., 1984; Maidak et al., 1994). The 16S rRNA
genes probably have greater potential than functional genes for providing the
means to evaluate AAO population sizes and structures within complex
communities.

Genes for enzymes in the autotrophic ammonia oxidation pathway
(McTavish et al., 1993; Sayavedra-Soto et al., 1994; Klotz and Norton, 1995)
may not provide sufficient target within community DNA to bind functional
probes, because AAO populations typically comprise less than 0.01% of the
total microbial community in soils (Schmidt and Belser, 1982). Functional gene

copy numbers and nucleotide sequences appear to vary enough within this
75

76

narrow group of B-Proteobacteria so that functional probe hybridizations may
only be able to provide relative estimates of AAO population sizes (Norton et al.,
1996; Bruns et al., manuscript submitted). Stephen at al. (1996a) recently
compiled a database containing 110 partial 168 rDNA sequences for AAOs in
the B-subgroup of Proteobacteria. Six distinct sequence clusters, based on
differences in variable regions of these sequences, were identified (Stephen et
al., 1996a). Marker sequences identified for each cluster appeared to be useful
for evaluating the composition of mixed AAO populations (Stephen at al.,
1996b).

Our study describes the use of PCR-based 16S rDNA analysis to
compare AAO populations in soils with different disturbance and fertilization
regimes. In contrast to workers who have used universal 16S rDNA primers
with PCR to characterize whole bacterial communities in simpler systems
(Giovannoni et al., 1990; Schmidt et al., 1991) we used more specific primers
(McCaig et al., 1994; Kowalchuk et al., 1996) targetted for AAOs and their close
relatives within the larger soil microbial community. We also analyzed PCR
amplification products with oligonucleotide probes developed from the
database of Stephen et al. (1996a). Two approaches were used: 1) cloning of
PCR amplification products followed by probe hybridizations of clone blots; and
2) separation of PCR products by denaturing gradient gel electrophoresis
(DGGE; Muyzer et al., 1993), followed by probe hybridizations of DGGE blots
(Kowalchuk et al., 1996; Stephen at al., 1996b). We observed reproducible
differences in 16S rDNA sequences generated from the soil community DNA
samples and attempted to relate these observations to known differences in soil
treatments (i.e., disturbance and fertilization histories). These soils were shown
to have different AAO population sizes and different nitrification capacities

(Bruns et al., manuscript in preparation). We propose that the differences in

77
rDNA sequences obtained from soil communities reflect actual differences

between in situ AAO populations within these communities.

MATERIALS AND METHODS

Long-Term Ecological Research site. Soils containing the AAO
populations to be analyzed were sampled from the Long-Term Ecological
Research (LTER) site at Kellogg Biological Station (KBS) in southwest Michigan
USA. (Robertson et al., 1996). This site was established in 1988 to study
ecological interactions affecting productivity, nutrient availability, and biotic
diversity in cropped, successional, and native ecosystems representative of the
Upper Midwest United States. Chemical, biological, and physical
measurements and agronomic parameters are taken regularly at this site during
each cropping season. Much of these data are accessible through the KBS
LTER home page on the World Wide Web (http://kbs.msu.edu/lter/). Some KBS
LTER data for relevant treatments, experimental plots, and sampling periods are
cited in this paper (see below). These provide supplemental information for
ecological interpretations of results from microbiological and molecular
analyses.

Soils and treatments. Soils from cultivated and never-tilled plots at
the LTER site were sampled in July 1993, july 1994, and August 1995. Soils
from successional plots were sampled in August 1995 only. Analysis of 16S
rDNA from soil bacterial community DNA was performed on 1994 and 1995 soil
samples obtained from two 1-ha replicate plots per treatment.

The cultivated plots (LTER Treatment 1) had been in a com-soybean
rotation from 1989 to 1994, with wheat introduced as a third rotation crop in
1995. This treatment was conventionally tilled (annual moldboard plowing,

disking, and cultivation), treated with prescribed applications of herbicides and

78
insecticides, and fertilized with ammonium nitrate. The N fertilizer was
broadcasted at a rate of 84 kg N per he to the corn crop and 56 kg N per he to
the wheat crop. Crops in this treatment were com, soybeans, and wheat in
1993, 1994, and 1995, respectively. The never-tilled plots (LTER Treatment 8)
had not been tilled after clearing of the native deciduous forest in 1959. (Most
of the vegetation in these plots now consists of C3 grasses.) Successional plots
(LTER Treatment 7) had been in com and soybeans for at least 40 years before
they were left in 1989 to revegetate with extant flora. The successional plots
were therefore under a disturbance regime intermediate between the cultivated
and never-tilled treatments. Nitrogen-fertilized micropIots (5m x 5m) were also
sampled to evaluate fertilization vs. disturbance effects. These micropIots
(designated here as Treatment 7F) received 125 kg N per ha broadcasted in
July of each year.

Soils at the LTER site are classified as Typic Hapludalfs (U.S. Soil
Classification System, 1992) belonging to the Kalamazoo and Oshtemo soil
series (fine, loamy, mixed, mesic). Soils were sampled to a 10-cm depth with a
2.5-cm corer (approximately 14 g fresh soil per core). The A horizons in these
scils are typically 20 cm deep. Soil samples in 1994 were composited from five
sampling stations in each of the two replicate plots per treatment. Samples in
1995 were composites of 20 soil cores taken near LTER sampling station #1 in
each plot (Robertson et al., 1996). All soil samples were stored at 4°C until they
were analyzed in the laboratory. Prior to subsampling, soils were mixed inside
plastic bags by manually kneading and shaking the bags. Gravel and other
debris were remoeved from these soils manually. Soil moisture contents were
determined from two 10-g subsamples dried at 110°C for 48 h. All results
reported here are based on soil dry weights.

79

Descriptions of each treatment and other biological and chemical
characteristics of soils from these treatments are summarized in Table 4.1.
Nitrate and ammonium measurements are made monthly during ice-free
conditions as part of routine KBS LTER site data collection (Robertson et al.,
1996). The nitrate and ammonium contents reported in Table 4.1 were
obtained from the KBS LTER Site Data Catalog on the World Wide Web
(htth/kbs.msu.edu/lter). These were determined on composited samples from
each replicate plot between April and November in 1994. Soil samples were
sieved (4 mm) and subsampied in triplicate. Subsamples (10 g fresh weight)
were extracted in 100 ml of 1 M KCI by shaking for one min, standing ovemight,
and reshaking prior to filtering with a type A/E 1-um pore size glass fiber filter.
Filtrates were analyzed on an Alpkem segmented flow analyzer (KBS LTER
Site Data Catalog, Soil inorganic Nitrogen, 1996).

Organic C, microbial biomass C, and direct microscopic counts reported
in Table 4.1 were determined from composited samples taken in July, 1994
from each replicate plot. Carbon analysis was performed on oven-dried,
ground samples in a Carlo Erba NA 1500 series 2 nitrogen/carbon analyzer
(Fisons Instruments, Beverly, Mass). Microbial biomass C contents were
measured by the chloroform-tumigation-incubation method (Horwath and Paul,
1995). Direct microscopic counts were made by staining soil smears with DTAF
[5-(4,6-dichlorotriazin-2-yl)amino fluorescein (Sigma Chemical Co., St. Louis,
Mo.)] and obtaining random digitized images of the smears under
epifluorescence microscopy (D. Harris, personal communication) with a charge-
coupled device camera (Princeton instruments, Trenton, N. J.). Images were
transferred to a Power Macintosh 7100/66 for counting as previously described
(Zhou et al., 1996). Most-probable number (MPN) counts of MOS were
determined for 1995 samples by blending soils in a Waring Blendor with 100

80

20:232.. 2. E28322... .._e .e 255 v

220.80.23.58 .2020... 6.5:

.o. moor one mom. 2. .0... 23.30. .2. 62:30:. 2923 2... 5.... 60.928 v .0 fig .33. .25.. can £833.... 352.. 23.335-20.898... 2.6.0.0200
.88 9:2. am: we. .8... .8822 2:885 _.em .8880 seo 2.... mm: mm... 882.88.. :. :ozo :8 28.: o...

.0. 2.0.3.5.. 28628.0, .3": 2...... .80... 0.3.32 2.... .0. 8:222:30E Ea... Be «one: .83 235362 522.... .22 E2. 2520... cox... 8:222:30: m
...8.§8...n:...e§e.._ .82. 8.2. 282. :o 8.... 2:2. mm: m9. .82 .888... 2:82.... mm: mm... .82 .8 .e 8.88...

 

 

8:25.... 82 1 oz
oz oz . .838 . . 8....“ oz... 0.2 . 28.888: 8.888
mo.x.o.o«. 3 v2.23... o... «Sign. a. 2.133.... «68:88...
_.8 o 8.. nos. 8 258 2......
.8358 8... 2:...
oz 02 mop x 8. we. mm me. x Cd“. od =8 o 6.. 358 3.203 82.0
meow
oz 8m :8 8. o 8.. 9.. o 8:59.. 3.88....
oz son. a.» an... e... an“. v...” «82 8880.22 9.8. a 8..
2.32 o... 2822 :82
oz 8...... to 6...“. n... .3... o... «88. 88.8.2.2 2.8 a 8..
2.802 9.. 2802 :82
8...... :o 2 8.. z 9.
82. coin mm .23.. Eco 05.2. 3.9!
2 8.. 8... 8:... 8.888.. 8.39:2... 83:8 5.... o: 8.. z 9. x. 88. 8:...
z 9. 8. .o 8.8.88 228.888 2.8: 5888.. 82888... 8.8.2 823288828

83:8 22.8.8. 2 8822.5 82.. 8.8: 82.. .8_ 88. 2122.828 .58....o...

8. .898 a 85.8.: 88. o. 8.... 2.8.. 9. :. 8.88 .8268... 8.2.. 88. o. 8.... 28..
8 .58... 28m 53: 8.82.8 8.8: 888.8: 8.82 9. .8... .e 8. 882.30 .88... 28:88:22

22888.... 8:28.888. 32.88:. .3828.

88.2.... n: .858... n .8582... o 85.8... . 85.8.: 8.2.2....

 

.38 8... use .88. .8... 88.8.. <8... 898.... .588 8.888 88.8.. .2.» mm: o... 8 285.8: 32.88: 2..
88.888: 882.8 :8: 8.8.8 3.8 .o 8.8.8.888 8.83... :8 8.8.8882. .3228... 8.8.8... 28.82:... .5. .39.:

81
mM phosphate buffer (pH 7.0) and making twofold dilutions in 96-weii microtiter
plates (Bruns et al., 1996). The wells in the microtiter plates contained ATCC
Medium 929 modified by reducing the (NH4)2SO4 concentration tenfold to 10
ppm (Bruns et al., manuscript in preparation). inoculated plates were sealed
with parafilm, wrapped in plastic bags, and incubated at 25°C in the dark for
two months before testing aliquots from the wells for nitrate and nitrite (Bruns et
al., manuscript in preparation).

DNA 'extrectlon from soils. Soil samples were stored at 4°C until
they were analyzed in the laboratory. Prior to subsampling, soils were mixed
inside plastic bags by manually kneading and shaking the bags. Gravel and
other debris were removed from the soils manually. Soil moisture contents
were determined from two 10-g subsamples dried at 110°C for 48 h. All results
reported here are based on soil dry weights. Soil community DNA was
extracted from 5-g subsamples (fresh weight) by the method of Zhou et ai.
(1996). DNA was separated from soil humic substances by subjecting the
entire crude extract to electrophoresis in 0.8% low-melting-point agarose (Gibco
BRL, Gaithersburg, Md). The DNA bands were excised from the gels, and the
agarose was dissolved with agarase (Boehringer Mannheim Corp.,
Indianapolis, Ind.). The DNA mixture was concentrated and washed twice with
distilled water in Centricon-1OO ultracentrifugal filters (Amicon, lnc., Beverly,
Mass). DNA concentrations and purities were determined at 260, 280, and 230
nm with a Hewlett Packard 8452A spectrophotometer (Hewlett Packard Co.,
Sunnyvale, Calif.).

Polymerase chain reaction with 165 rDNA primers. Purified
DNA was amplified using the 16S rDNA primers, described by McCaig et al.
(1994). These have a higher specificity than universal eubacterial primers for

M05 in the fl-subgroup of Proteobacteria. The sequence of the forward primer,

82

B-AMOf, was 5'-TGGGGRATAACGCAYCGAAAG-3‘, which corresponded to
positions 141-161 of the Eschen’chia coli rDNA molecule. The sequence of the
reverse primer, B-AMOr, was 5'-AGACTCCGATCCGGACTACG-a',
corresponding to E. coli positions 1301-1320 (McCaig et al., 1994). The PCR
was carried out in 50-ul reaction volumes with a Perkin-Elmer GeneAmp PCR
System 9600 (Perkin -Elmer, Foster City, Calif), using a hot-start procedure to
reduce nonspecific amplification (Chou et al., 1992). Each reaction mixture
contained 10 ng template DNA, 4 pmol of each primer, and 1 Unit of Taq
polymerase (Perkin Elmer, Foster City, Calif.) in a final concentration of 2.5 mM
MgCl2, 0.12 mM dNTPs in PCR reaction buffer. Positive controls contained 10
ng of Nitrosomonas europaea genomic DNA as template. Negative controls
contained dilutions of processed agarose gel slices that were cut from areas in
the purification gels containing no DNA. PCR reaction conditions were as
follows: initial denaturation at 94°C for 2 min; 35 cycles of 94°C for 1 min/55°C
for 1 min/72°C for 1 min; final extension at 72°C for 7 min. Four to five
duplicate reaction mixtures were amplified at a time, and the PCR products
were pooled to reduce bias resulting from PCR drift (Moyer et al., 1994).
Clonlng of PCR products. Pooled mixtures of PCR products were
concentrated in a Microcon-100 unit (Amicon, Inc., Beverly, Mass.) and added to
agarose gels for electrophoresis to isolate PCR product bands. Gel slices
containing the PCR products (approximately 1100 bp) were excised and treated
with QIAEX ll reagents (Qiagen, lnc, Chatsworth, Calif.) to purify the DNA from
the gel. Purified DNA was washed again and concentrated in a Microcon-100
unit. PCR products were ligated into the pGEM-T vector using T4 ligase
(Promega, Inc., Madison, Wisc.). Epicurean coli XL1-Blue supercompetent cells
(Stratagene, lnc., La Jolla, Calif.) were prepared and transformed with the

ligation mixtures according to the manufacturer's directions. The entire

83
transformation mixtures were plated out onto Luria-Broth (LB) agar plates
containing 20 mg ampicillin, 80 mg methicillin, and 10 mg kanamycin per liter of
agar. (Methicillin was added to reduce incidence of false-positive satellite
colonies.) Immediately prior to inoculation, plates had been freshly prepared
and spread with 50 ul IPTG solution (0.2 M isopropyl-f-thio-B-D-galactoside in
water) and 50 ul Xgal solution (20 mg 5-bromo-4-chloro-3-indolyl-B-D-
galactoside per ml of dimethylformamide) to permit chromogenic detection of
white transformant colonies from blue nontranformant colonies. Plates were
incubated at 37°C for 24 h, then held at 4°C for 48 h to allow nontransformant
colonies to develop deeper blue color. Well-isolated, white colonies were
picked to fresh LB plates containing 100 pg ampicillin per ml to produce a 168
rDNA clone library for each soil sample. Plasmid DNA preparations were
obtained from these clones using the Wizard Minipreps kit (Promega Corp.,
Madison, Wisc.).

T-tracklng and sequencing of PCR Inserts. Previous observations
during development of the database by Stephen et al. (1996a) had shown that
the primers of McCaig et al. (1994) could generate nonpecific PCR amplification
products from non-AAO DNA (e.g., DNA from Comamonas festosteroni and
some other related B—Proteobacteria). T-tracking was used to distinguish AAO
rDNA inserts from non-AAO inserts. (With the T-tracking method, PCR inserts
are partially sequenced with a single terminating dideoxynucleotide (dATP) to
generate single-lane band patterns in sequencing gels.) The T-track band
patterns were generated from the 168 rDNA region corresponding to positions
300-500 of E. coli rDNA. T-tracks from AAO rDNA had characteristic motifs that
were used to screen out clones containing non-AAO rDNA (Stephen at al.,

1996a). In this study, T-tracks for 45 clones were obtained from plasmid DNA

84

samples with the sear 168 rDNA sequencing primer, 35S-dATP, and a
Sequenase kit (Amersham Corp., Arlington Heights, Ill.).

Plasmid DNA samples, which generated different T-tracks having the
characteristic AAO motif, were sequenced manually across the 168 rDNA
region corresponding to E. coli positions 200-530. Sequences obtained from
these plasmids were aligned using the GDE program (Maidak et al., 1994) with
other partial 168 rDNA AAO sequences in the database of Stephen at al.
(1996). Phylogenetic analysis (Jukes and Cantor, 1969) was performed using
PAUP (Swofford, 1993) and Phylip (Felsenstein, 1993) to generate a
phylogenetic tree based on nearest-neighbor distances (Saitou and Nei, 1987).
GenBank accession numbers for these sequences are U56606 to U56633.
Three plasmid DNA samples that generated T-tracks with non-AAO motifs were
also sequenced to confirm their positions in the phylogenetic trees.

Colony blots and probe hybrldlzatlons. Clones were transferred
as 3- to 4-mm patches onto LB-ampicillin plates and incubated overnight at
37°C. Growth from these plates was transferred to replicate 10 cm x 10 cm
plates (100 clones per plate) and incubated at 37°C for 18-20 h. Colony
transfers were made from each plate by applying a Hybond N+ membrane
(Amersham Corp., Arlington Heights, III.) to the surface of a plate and leaan it
for one min. The membrane was then lifted from the plate and placed colony-
side-up on a blotting paper soaked with denaturing solution (1.5 M NaCl, 0.5 M
NaOH) for 7 min to lyse the cells. The membrane was neutralized (1.5 M NaCl,
1 M Tris, pH 7.4) and rinsed in 2X saline-sodium citrate (SSC) buffer before air-
drying. The DNA from the clones was cross-linked to the membrane by UV
irradiation in a Stratalinker (Stratagene, lnc., La Jolla, Calif). Clones containing

plasmids with PCR inserts with known sequences were included in the clone

85
libraries during membrane preparation to serve as positive and negative
hybridization controls.

The sequence database of Stephen at al. (1996)a was used to identify
oligonucleotides that could be used as probes (T able 4.2) to differentiate
various subgroups of AAO sequences in the database. Probes were end-
labeled with g32P-ATP (DuPont NEN Biotechnology Division, Wilmington, Del.)
with T4 polynucleotide kinase (Stratagene, Inc., La Jolla, Calif). The 32P-
labelled probes were added to hybridization solution to obtain activities of
approximately 1 mCi per ml.

Membranes containing DNA from clone libraries were prehybridized in
QuikHyb solution (Stratagene, Inc., La Jolla, Calif.) for 1 to 2 h in glass
hybridization tubes in a Techne Hybridization oven (T echne, lnc.. Princeton,
N.J.). Membranes were hybridized with 32P-labelled probes at 42°C for 6 to 18
h. Membranes were washed at 42°C in 2X SSC containing 0.1% sodium
dodecyl sulfate (SDS) and placed in film cassettes for exposure to
autoradiogram film (Kodak, lnc., Rochester, N.Y.).

Denaturlng gradient gel electrophoresis (DGGE). Purified DNA
extracts (20 ng per reaction) were used in PCR amplification reactions with the
primers CTO178F-GC and CT0637r (E. coli numbering), which were developed
specifically for DGGE analysis of AAO populations by Kowalchuk et al. (1996).
This primer pair generates a 497-bp fragment containing 459 bp of rDNA
sequdice and a 38-bp GC clamp. PCR reactions were carried out using the
Expand High Fidelity PCR System (Boehringer-Mannheim Corp., Mannheim,
Germany) in 50-ul volumes overlaid with an equal volume of paraffin oil. PCR
conditions in an OmniGene Thermal Cycler were as follows: Initial denaturation
at 94°C for 1 min; 35 cycles of denaturation (92°C for 30 sec)/annealing (57°C

for 1 min)/extension (68°C for 45 sec with an incremental increase of 1 sec per

86

.0920 00:00.30 :0 .ozP

 

565...:

3.0.5....
.0020 =3, .05 $8.0 mom
2:0 0. 3.25: 0265002.?

35:20. 0:02

ounce...
285 oom c. 88808 oz

3.22...

.83 ___.s .2. 52.0 on...
28 0. 03:2 89:82....

.505...
239.0 on... c. 88308 oz
355.. 952

28.5.... 028.28.... .22o
2.85. 22.2... .n. 3&0 m

..EE0009§00 039:8:00Q

F0<h0<h000<0h<<h0

G<<<0h0<00h0t0E

<<h0<OOOEOEOFO

0t00EOO<<F000<

000<Gh0000<t<h06<

0h000<<t<t0<0<0b

G<<G 06000.:on

0h<0<0<0h<<h00<h0<

0h<0h<00020<t<0
€00.00...
<<<0<<0<00h0<0E0
.0900...
0<0<<<0<<<oooho<t

8:00...
P000<t00<<<<00<<0

.0300...
000520000<t§00<

.613.
<055<§<<t000<0
€9.00...
300500000<0E0

800.000
p0<p00<t<055~<o

0022850050..
.8230 c305. o: .050... 00:02.00
0522228.. 5.; 88.0 8.5...

.3923 3.02.0. 0E8 ES.

00550. 230300.. 0.0 .05 00.0.3. not
5:00.95 0022:8903. 00.3.38: 61020.2
fox 5 0005.600 8|
:0. 00.658085 00.2.30 6109.02

.cxo. 000. 022.30 0.2. .5

02:00an 30: 052:0 00.3.3.5
282.5... 238$... .262 30.220

mom $0005.63 83000222
Eo... «05.0.0 .3 o. 3......»
08:308. gong... 02230 01.0.2.8

$0.350 3502.2 2:00 50...
.85.! 25:02. so .2. 8.28.

 

£032.-»er

.0 o. .0. 08.58 ..8

:00 :02? 0:330 $98.52 09.03000 one... 00co0._E< .m 0:0 0053000 (20. m9

5:832. 2%822 82.32: 30.220
2188...: .< 220...?

.838. 8.2.8 .5
02500.00. .30 03920025,. 0:0 I 10303
2.08822. 8.} 323:0... .2 _0 2.5
.3...
0:06 .090... one...

 

.88. ...o .o 5.8.0. 2382. 20.0.5352...

8.8.850 9233-.. 22. e883?» <20. 3.2. o: .o 283.8 o... 22. .8528. 88803 32.. 8.38825 .3. 0.55

87

cycle); and final extension at 68°C for 5 min (Stephen at al.. 1996a). DGGE
was carried out according to the methods of Muyzer et al., (1993) in 20x20 cm,
1.5-mm-thick, 0.8% polyacrylamide gradient gels (0.5x TAE; 37:1
acrylamide/bisacrylamide) with a D-Gene system (BiocFlad Laboratories.
Hercules, Calif). (1X TAE contains 0.04 M Tris base, 0.02 M sodium acetate,
and 1 mM EDTA, pH 7.4). Gels were poured with a CBS gradientmaker (Del
Mar, USA) and a Bio-Rad Econo-pump (Bio-Rad Laboratories, ) at a speed of 5
ml per min (40-ml gradient volume) to obtain a gradient of typically 38 to 50%
denaturant. DGGE gels were run in 0.5x TAE at 60°C for either 6 1/2 h at 200 V
or 16 h at 65 V. Gels were stained with ethidium bromide to visualize DNA
bands under UV transillumination. Southern blots were prepared from the gels,
and these were hybridized with selected 32P-labelled probes (Table 4.2).

RESULTS

DNA yields from soils. DNA yields from cultivated soils were
significantly lower (p < 0.05) than yields from never-tilled soils in 1993 and 1994
(Table 4.3). No significant differences were observed between DNA yields from
cultivated, never-tilled, and successional soils in 1995. DNA quality, as
measured by optical density (0. D.) ratios of DNA preparations at 260 nm/280
nm, was slightly higher for 1994 samples (1.8 to 1.9) than for 1995 samples (1.5
to 1.7). The O. 0.250/230 ratios for the same samples were also higher in 1994
(2.3 to 4.4) than in 1995 (1.3 to 1.5). Since the latter ratio is a measure of humic
substance contamination in nucleic acid samples, the 1995 DNA samples
were less pure than the 1994 samples. Both sets of samples, however,
appeared to yield equivalent PCR product band intensities when amplified at 10
ng DNA per 50—ul reaction.

TABLE 4.3. DNA yields (ug DNA per 9 soil dry weight) from cultivated, never-
tilled, and successional soils. Numbers in parentheses indicate ranges of
experimental values for each treatment.

 

 

Treatment 19931 19942 19952

(corn) (soybean) (wheat)
Cultivated (T1) 2.6 (11.4) 2.9 (10.1) ) 12.82 (13.8)
Never -tilled (T 8) 15.4 ($3.3) 13.0 (12.8) ) 15.7 (13.7)
Successional (T7) ND ND 13.6 (11.4)
Fertilized successional ND ND 19.6 (13.7)
(17F)

 

1 Mean of three analytical replicate subsamples from 1 composited sample for
each treatment.

2 Mean 012 replicate samples from each treatment.

89

Sequence diversity in clone llbrarles. Twenty-six clones obtained
from community DNA of the never-tilled soils sampled in 1994 were screened
by T-tracking. Half of these (13 clones) exhibited characteristic AAO motifs.
Nearly all T-tracked clones (18 out of 19) from community DNA of cultivated
scils exhibited AAO motifs (Figure 4.1). The AAO motif was characterized by the
sequence TCTTI' at E. coli positions 432-436, shown by the bracketed region in
Figure 4.1. T-tracks that did not have this motif (tracks 2, 18, 21, 22, 24, 25, 26,
28. 29, 31, 34, and 35) were considered to represent nonspecific PCR
amplification products (cloned DNA that did not arise from AAO rDNA). From
the percentage of T-tracks with AAO motifs in this sample set of clones, we
concluded that clone libraries from community DNA of never-tilled soils
contained proportionally more clones with nonpecific amplification products
than libraries from cultivated soil community DNA. These T-tracks also provided
information on comparative diversity within this sample of cloned rDNA
sequences. One out of 13 clones containing the AAO motif in the never-tilled
sample set had a redundant T-track pattern (similar to that of another clone in
the set). Six out of 18 (one in three) clones with AAO motifs had redundant
patterns. Although the sample set of clones from never-tilled soil community
DNA contained proportionally fewer A0 rDNA inserts, fourfold more clones had
unique T-track patterns than did clones in the cultivated soil sample set.

Phylogenetic analysis was performed on 24 cloned sequences that had
been aligned to other AAO sequences in the database of Stephen at al., 1996a.
A phylogenetic tree was constructed (Figure 4.2) containing 13 sequences from
cultivated soil community libraries and 11 from never-tilled soil community
libraries. All sequences from cultivated soil community libraries fell within one
clade (Cluster 3 Nitrosospira) in the phylogenetic tree (Figure 4.2). The most

dissimilar AAO sequences among these clones differed at 3% of the base

 

 

 

 

    
 

 

 

  

- .—
. a“?
‘ ,_., , : . - .1
"' + ' i ’5’ f, 942;, ‘g
'!.t~ i 'V
a. “i 3%: s . 2! ' " -
5'32 1,; " .. V. '
L ‘ w
.} v‘ ‘
a 3 . .
~ - rm: . "= 1.5-g.
' f‘“%; f - a l
i ' i=3} i ‘ finnr .
"it“ 4 ' ;’ if ‘ -q
. : whim-‘7: ' ‘ii‘VL' . -fl
- , —‘ ' :13 - ‘ ~-
,. ‘ a I“, :, . . i: -, fl
1 . i ".s ~r~¥i€g3 s? ‘ 1'
.- ' a}; ‘1 .3”- ’ “
:E‘L'“ ‘ in. El: .7 > ‘ J: -.‘;;" .27 ‘I
‘5’ , _ -::;1 i {'51. I. :f‘; 5.. “i ' if
or. . “I. ' —_ ‘
s»
> . 2 .. w . N. r u ‘ " ~ ‘ r
. ’ - w. ‘ .- 5:" ~22. .:3 - .J".
‘ tii-m‘; ” a ,

+-+++++++++++++++ -++--+-——+--OO+-++--

”girl:
3 T.

l‘ " l 1-

   

FIGURE 4.1. T-track banding patterns observed with single-dideoxynucleotide termination
sequencing reactions. DNA samples sequenced were partial 16$ rDNA gene inserts obtained
from PCR amplification of soil community DNA with the b-AMO primers of McCaig et al.. (1994).
The left-hand set of 17 banding patterns (T -tracks 1-17) were sequences amplified from cultivated
soil community DNA. Ali T-track patterns, except one. in this set, contained the characteristic AAO
motif (see indicator arrows) composed of a singlet band, a singlet gap, and a triplet of bands. The
right-hand set of 18 band patterns (T -tracks 18-35, not including the two inner lanes with no DNA)
were sequences amplified from never-tilled soil community DNA. Seven of the 18 T-track patterns
contained the AAO motif, while 11 patterns appeared to be derived from non-AAO DNA. (Lanes
with the AAO motif are marked "+'; lanes without this motif are marked '-'.)

91

FIGURE 4.2. Phylogenetic tree showing distances between 168
rDNA sequences obtained from cultivated soil samples (KZOO-H
sequences)and never-tilled soil samples (KZOO-D sequences).
Phylogenetic analysis included other partial 16$ rDNA sequences

for B-subgroup autotrophic ammonia oxidizers from the sequence
database of Stephen et al. (1996a). Clusters indicated are
phylogenetically distinct groups of sequences identified in this
database. Sources of other sequences in the tree were: neutral
Craibstone (Scotland) soils (PH7); acid Craibstone soils (PH4.5
and 4.5); marine waters and sediments (Env and JUNE);
low-nitrogen-ioaded activated sludge (AL212 from Y. Suwa);

and enrichments from pig waste slurries (JWU sequences from

A. Princic).

 

 

 

Nitrosococcus ooeanus

92

Escherichia coli Non-AAOs __

 

 

 

t .
Rhodocyclus. enuis K 026

 

 

 

 

 

 

 

 

 

 

H4. SAIGZ“
firm 5,0022: Cluster 2 Nitrosospira,
p H7C/54‘

ifl m5) $200315

£1.1ch

 

 

 

'WW Deep-branched
2

 

 

  

nv 1—
5.65-2133 Cluster 1 Nitrosospira

r996)-
.“4'5A’D‘ Cluster 4 Nitrosospira
4.5 E“

 

i
z‘é%°€rzt?25 ——J-

rtros c us multiformis
K200 _H16

15523132353

31
Nitrosospirabricnsisstr.C-128
KZOO_ H118

Nitzrososlsiiha sp. str. NP22- 21
KZ'OO_ H7

K126031813; Cluster 3 Nitrosospira
K200 H10

HZOOO'O HM
iii” a
K200 _H6

 

  
  
 

Mu

 

 

   
  
 

hfirgggsgi‘rasp. str. NpAV
Nitrosovrbriotcnuis str. Nvi

K H5
“will? N2

I
pins 3' _..i

KZOO_D27
fiimfimm
PH7B ‘
C137“,Z2

ALZ 2
BNiztrosococcus mobiiis str. Né

JUNEI4
9K83 _JWU91

—-l E w‘l?“
871C 1W 7
n NJWU88 8B!" _JW0889
—L81\'itrosNomonas cutropha str. C-91
EnvA 135158 itrosomonas europaea

gifizzéziZCluster 6 Nitrosomonas
nvA

 

 

 

 

 

 

 

J NEE/11
AzoarcuTnitrificans str. Toi-4

 

 

Alcaiigenes cutrophus

 

Gnilionclla ferruginea

 

 

FIGURE 4 . 2

Mcthyiophilus methyiotrophus

Chromauum vrnosm

0.10

 

93
positions (out of a total 318 bases). Of the 11 sequences from libraries of never-
tilied soil community DNA, three fell within Nitrosospira Cluster 3, two within
Nitrosospira Cluster 5, one within the Nitrosomonas clade (Cluster 6), and six
within a deep-branching clade having intermediate phylogenetic distance
between the Nitrosomonas and Nitrosospira groups. The three Cluster 3
Nitrosospira sequences were identical, so these were counted as one distinct
sequence to give a total of 10 different sequences from never-tilled soil
community libraries (Table 4.4). The most dissimilar AAO sequences derived
from never-tilled soil community DNA differed at 14% of the 318 base positions.
Based on this comparison, AAO sequences from never-tilled soil community
libraries exhibited fourfold greater diversity than AAO sequences from cultivated
soil community libraries.

Probe hybridizations of clone-blots. Three oligonucleotide probe
sequences (Table 4.2), developed from the database of Stephen at al. (1996a),
were used in hybridization tests with clone-blots to screen larger numbers of
clones for non-AAO PCR products. Based on information in this database, the
Ammo_Cl_2/3/4/6 probe could hybridize to all terrestrial AAO sequences,
except for sequences falling within the deep-branched clade. The B-Deep and
All_Spira probes could hybridize to sequences in the deep-branched clade and
all-Nitrosospira group, respectively. The percentages of clones with DNA
hybridizing to the Ammo_Cl_2/3/4/6 probe were higher in libraries from
cultivated soil community DNA than in never-tilled soil community libraries
(Table 4.5). Percentages of clones cross-hybridizing with the B-Deep probe
were higher in never-tilled soil community libraries than in cultivated soil
community libraries.

Probe hybridizations with clone libraries derived from soils sampled in

1995 were performed to determine whether similar hybridization percentages

94

TABLE 4.4. Nunbers of diferent partial 16$ rDNA sequences falling within sibgrotps in B-
Proteobacteria (1994 sanples).

 

 

Treatment Total Typical Novel Nitrosomonas Deep-
number of Nitrosospira Nitrosospira clade clade branched
sequences clade clade

Cultivated 1 3 1 3 0 0 0

Never- 1 0 1 2 1 6

filled

 

95

could be obtained in a subsequent year. In addition, clone libraries were
obtained from soils sampled from successional plots that had a disturbance
history intermediate between those of never-tilled and cultivated plots. The
degree of nonspecific amplification from community DNA of 1995 soils (84 to
99%) was much higher than that from community DNA of 1994 soils (60 to 80%
in Table 4.5). The number of clones hybridizing to the Ammo_Cl_2/3/4/6 probe
were higher in libraries from cultivated and fertilized successional plots than for
never-tilled and successional plots (Table 4.6). Numbers of clones hybridizing
to the B-Deep probe were highest in libraries from never-cultivated soils.

Denaturing gradient gel electrophoresis (DGGE). PCR products
were obtained from all soil samples using DGGE primers with greater specificity
for AAOs (Kowalchuk et al., 1996). The DGGE patterns for replicate samples
within each treatment were remarkably similar, except for soils from fertilized
successional plots (Figure 4.3). Very similar DGGE patterns were also
observed in samples from the same plots taken in 1994 and 1995. Bands 1 and
2 in lanes containing DNA from cultivated soils were more intense than
corresponding bands at these positions in lanes containing DNA from never-
tilied soils. These same bands appeared stronger in lanes containing DNA
from successional soil samples. Their intensities, however, were not as great
as they were in lanes with DNA from cultivated soil samples. The third band
observed in all lanes containing DNA from cultivated soil samples was not seen
in lanes containing DNA from never-tilled soil samples. Conversely, the third
band in lanes containing DNA from never-tilled soil samples was not observed
in any of the lanes containing DNA from cultivated soil samples.

DNA in DGGE gels was transferred to a hybridization membrane and
probed with the All_Spira probe, which hybridizes with all Nitrosospira
sequences in the database of Stephen at al. (1996a). Equivalent hybridization

TABLE 4.5. Percentages of Win clones in lbraries obtained from soil community DNA
(1994 samples).

 

Treatment Replicate Total Percentage Percentage positive Percentage positive
plot clones negative for al fortypical and novel for deep-branched

 

tested probes AAOs clade probe
Cukivated 1 81 36% 64% 0%
2 200 44% 56% 0%
Never- 1 346 78% 10% 1 2%
filled 2 326 79% 4% 1 7%

 

97

TABLE4.6. Numbersdprcbe—posihrecbneskilbrafiesobtahedfromsoioomuninNA
(1995 samples).

Treatment Replicate Total Number hybridizing with Number hybridizing with

 

 

plot clones all-AAO probe deep-branched probe
tested

Cultivated 1 500 73 6

2 494 4 0
Never-tilled 1 482 3 1 5

2 480 3 46
Successional 1 369 2 2

2 457 7 2
Fertilized 1 396 32 9
successional 2 486 1 0 3

 

98

.30. :. 00.0805 :0 55; 5.56.05... 55.550030 .m 00. 5:05.583 50.22:: M; 0:3 .0 00. 55.05830

:. 5.005.... 002.5. .5 55.. S 05.. 55.50830 53.2.5.5 .0. 5.5.. u. 05.. 55.80005 :. 8.005.... 005.25. .0 05.. £8. .0 00. .00....
.55: .w 0.5.. .30.. .v 5.. 05....-55: K 55.. .39 .m 05. 05.....55: .m 55.. .39 .m 05.. 00.....55: .m 0:5... .30. .o 5.. 05.02.50
:0 55.. .30. .w .5. 05.02.50 .m 05.. 68. .m 5. 00.02.30 .N 05.. .50.. .m 05. 00.02.50 .. 0:0... "€050.05 mm 050.. 9.3605

0:0 5.0... 0.00.39. 30:0. 50 .mm m... 0.:. .5000. 55; 53:50 .68 5 .0 .5 0.35.0302 .0 n5EE. 00010.5 5.; <20 2.:3EE00 =05

.0 :0..00....0Em mun. 5 550.00 55.2.: E9. 30 B... 0.5:..98. <20. mm. .5550. 50:00 <20 .:0..m:_E:_.. >0 50:: 5.03 .5605
0:55:50 00:.0.m-00.:.05-E3.0.:5 :. 002800 <20 2.:3E500 E0... 0.03005 :0..00..._..Em-mon. .0 5:5qu 0:00 0000 .m... mung".

   

   
  

.230 .230 .230 .230

 

003m 003m 003m 003m .2052 .052 .052 2552

99
signals were observed from banding patterns from all samples. When the blots
were probed with the Spira_Cl_3 probe (for Cluster 3 Nitrosospira). strong
hybridization signals were obtained in lanes containing DNA from cultivated
and fertilized successional soil (Figure 4.4). No signal was observed between
the Spira_Cl_3 probe and DNA from both replicate plots of never-tilled soils.
Furthermore, no signal was observed in the lane containing DNA from one
unfertilized successional replicate plot. When blots were probed with the
All_Spira probe, all bands from all samples had exhibited approximately
equivalent hybridization intensities (Figure 4.5). Hybridization patterns in
Figures 4.4 and 4.5 indicated that bands from community DNA of never-tilled
soils and unfertilized successional soils were derived from Nitrosospira DNA

but specifically not from Cluster 3 Nitrosospira.

DISCUSSION

This study showed that AAO ribosomal gene sequences from never-tilled
soils exhibited fourfold greater dissimilarity than AAO sequences from cultivated
soils. These results provide strong evidence that AAO populations were more
heterogeneous in the never-tilled soils than in the cultivated soils. Extraction of
DNA from soils, PCR amplification, or cloning can cause biased sampling of in
situ ribosomal gene sequences (Giovannoni et al., 1996). However, the fact
that these samples were analyzed with the same procedures under identical
conditions strengthens the argument that observed sequence differences
reflected actual differences in the in situ AAO populations. A reasonable
explanation for these differences is that AAO diversity is reduced in cultivated
soils by a selection process resulting from regular additions of N fertilizers.
Lower AAO diversity in cultivated soils could also be caused by a reduction in

niche heterogeneity due to regular plowing disturbance.

100

.000. :. 00.00.00 :0 0.0; 0.8.502. 08.000005 .0 00. 08.000800 000.0020 .N. 0:0. .0 00. 08.80800 :. 8.00.0.5 000.05. .2 0:0.

.. 00. 08.8083 000.0020 .0. 0:0. .. 00. 08.00083 :. 0.00.02. 03.....0. .0 0:0. .000. .0 00. 00......05: .0 0:0. .000. .v 00. 00....-.05: K 000.. .000.
.m 00. 00.....55: .0 0:0. .000. .m 00.. 00....-55: .m 000.. .08.. .o 00. 00.02.50 .0 0:0. .30. .0 00. 00.02.30 .m 0:... $.00. .m 00. 00.02200 .N 0:0. .30. .m
:0. 00.02.30 .. 000-. ”Em“... 00:00.00 0:0 0.0.0 0.00.30. 5.5 2.00 0532.0. 0... ES. 00.00.06 <20 0.5.5.80 E2. 00....0E0 0.55%.. 0:00 .0 80000000....
000:. >>0:0 00:3 to... 0.00.”. :_ .00 80.. <20 030:0 .20 .000... 01.0-2.8 00:80.00. 5.3 0000.0 .0... 0000 .0 5.00.0805... .0... 0000.”.

 

mm mm mm mm mm em mm vm mm 00 mm em
22:: to... 20...: to". eméh wméh 0:0... ug.-o... 2.7.-h mm._..r mm-_:_. mm-w._.
mCi-h 2..-»... wmihh «mink 00:: 02:. 00:: 00:: .023 .023 no; to;
003m 0030 0030 0030 .o>oz .052 .052 .052 $530 $230 .>...30 ->_..30

101

.000. :. 00.0500 ..0 0.0.5 0.:0E.00..

.0:o.00083w .m 00.. .0:o.000oo:0 00N.._t0.:: .m— 0:0-. .0 00.. .0co.000oo=0 :. 6.092.: 005.00. .5 0:0... 3 a0: .0:o.000oo=0 00N...00.::

.9. 0:04 2. a0. 03.000830 :_ .o_no.o_E 005.00. .0 0:0-. .000. .v 00.. 00....-.0>0: .0 0:0-. .000. .0 :0. 00....-.0>0: K 0:0. .000. .0

Q0: 00....-.0>0: .0 0:0. .000. .0 a0. 00....-.0>0: .0 0:0-_ 600. .0 Q0: 00.02.50 .v 0:0. #00— .0 :0. 00.02.50 .0 0:0-. .000. .m 00. 00.02.50
.N 0:0. .000. .m 00. 00.02.30 .F 0:04 ”0.60 0526.6. 0:. E2. 00.00.00 (Zn. b.5588 E2. 00....0E0 0.:0Eo0... 5.3 05005000.... 0005
26:0 00:0. .00 0.30.“. :. .00 Eo... <20 039.0 SE .000... 0..qwl..< 00..0n0.-n.mm 5.3 00005 .0... moan. .o E0..mo.00.o.:< .m.v mmDOE

 

mm mm 00 mm mm 00 00 0.0 00 em 00 00
00...: to". 00...... :0". 3.0.. 0.0.0.. 2.0.. 0...: 0:. Fr 0:. E. 0....- E. m...- r.-
nm-t 0a.: FEL- E-p 00.... 00.... 00.... 00.... 00.0> 00.3 00.2, 00.:

83m 003w 025 026 -.0>0z L052 .332 L302 -030 -050 -.._:o .030

1 02

Another important finding was that Nitrosospira Cluster 3 sequences
predominated in DGGE blots from cultivated and fertilized successional soils,
whereas they were not detected in blots from never-tilled and successional soils
(Figures 4.3 and 4.4). Nitrosospira Cluster 3 is one among seven
phylogenetically distinct B-AAO clusters identified by Stephen at al. (1996a and
1996b) from a sequence database derived from physically and geographically
diverse samples. The Nitrosospira and Nitrosomonas groups in this database
comprise four and three clusters, respectively. The Nitrosospira group consists
of Cluster 1 (sequences from marine samples), Cluster 2 (sequences frequently
retrieved from soils but without any cultured representatives), Cluster 3
(sequences of Nitrosospira spp. commonly cultured from soils), and Cluster 4
(novel sequences from previously uncultured Nitrosospira spp. with pure
cultures recently isolated; Stephen et al., 1996a). Nitrosospira Cluster 3
includes cultures and sequences that were commonly obtained from
agricultural soils in Scotland and the Netherlands (Stephen at al., 1996a). The
presence of Cluster 3 sequences in fertilized soils and the lack of these
sequences in never-tilled and successional soils suggest that this cluster is
associated with soils that have been recently amended with N fertilizers.

Nitrosospira sequences other than Cluster 3 sequences predominated in
DGGE mixtures from never-tilled and successional soils, because DGGE blots
from these samples hybridized only with the All_Spira probe (Figure 4.4).
Since it was not likely that soils contained Cluster 1 Nitrosospira (marine
strains), these DGGE mixtures probably contained Cluster 2 or 4 Nitrosospira.
These clusters appear to include sequences from less readily culturable
Nitrosospira spp. No representatives of Cluster 2 sequences have been
isolated in culture, whereas a few Cluster 4 isolates have recently been

obtained (J. l. Prosser, personal communication). Bruns et al. (1996) isolated a

103
representative of Cluster 4 Nitrosospira from a never-tilled sample after a six-
month enrichment in mineral medium containing 10 ppm NH4-N. This NH4—N
concentration, which is tenfold lower than concentrations typically used for
most-probable-number (MPN) enumerations of MOS (Schmidt and Belser,
1982), was an important factor in the successful isolation of Nitrosospira strain
Flt, a novel representative of Cluster 4 Nitrosospira (Bruns et al., manuscript in
preparation). Other workers have used lower NH4-N concentrations in MPN
media to increase AAO recovery from forest soils (Donaldson and Henderson,
1989). This may indicate that the most commonly used methods to culture
AAOs have been inappropriate for recovering certain AAO groups.
Predominance of Cluster 4 sequences from never-tilled and successional soils
may shed light on previous studies, in which no or low numbers of nitrifiers were
recovered from grassland, forest, and climax ecosystem soils that exhibited
nitrification after the vegetation was removed (Rice and Pancholy, etc.). These
observations led some researchers to suggest that certain plants inhibit nitrifier
pOpulations (Rice and Pancholy, 1978). An alternative explanation, based on
our sequence observations, is that the predominant nitrifiers in these grassland,
forest, and climax soils were not culturable in recovery media used in these
studies.

Another difference observed in our study was that sequences from a
deeply branched cluster of unidentified B-proteobacteria appeared to be more
common in clone libraries from never-tilled soils than in libraries from cultivated
soils. This difference in frequencies of deep-branched sequences was
observed in both replicates of each treatment in 1994 and 1995 (T ables 4.5 and
4.6). The deep-branched sequences appeared to be derived from bacteria
which are related to AAOs. These bacteria cannot be designated as AAOs until

cultured representatives have been shown to oxidize ammonia autotrophically.

104
Deep-branched sequences have also been retrieved from Scottish soils,
sediments, and freshwater samples using the B-AMO primers of McCaig et al.
(1994). The CTO_PCR primers used for DGGE (Kowalchuk et al., 1996) were
more specific for AAOs than the B-AMO primers (McCaig et al., 1994), and they
did not amplify DNA from plasmids containing inserted sequences from the
deep-branched cluster (unpublished observations). Thus. the DGGE blots in
our study were not expected to contain any deep-branched sequences. The
information on this cluster was available only from the clone library analysis and
not from DGGE results. The higher frequency of deep-branched sequences in
clone libraries appears to be associated with less disturbed soils, but the
phylogenetic and functional significance of this association is unknown. Further
assessments for the presence of these sequences in other environmental
samples and successful isolation of deep-branched representatives are needed
to learn more about the function and significance of these bacteria in soils. If
their association with undisturbed soils is found to be robust, these deep-
branched B-Proteobacterial sequences might potentially serve as disturbance
indicators in soil restoration programs.

A final important finding was the close similarity between DGGE band
patterns for PCR mixtures from replicate plots of the same treatments from both
years (Figure 4.2). All replicate plots had an area of 1 ha and were located
among a total of 46 plots in a randomized block design over a 48-ha site
(Robertson et al., 1996). Variability of soil properties at this site (including net
nitrification) was observed to be high and spatially dependent over distances of
1 to 40 m (Robertson et al., 1993). Approximate distances between sampled
replicate plots in our study were 900 m, 500 m, and 100 m, for cultivated,
successional, and never-tilled treatments, respectively. Nearly identical DGGE

patterns were obtained with each sample that had been composited from 25

105

cores taken from different sites in each plot. At least two explanations can be
given for DGGE pattern reproducibility. First, reproducibility could result from
AAO populations in these soils being truly robust and strongly dependent on
disturbance history. Alternatively, reproducibility could be due to nucleic acid
selectivity of PCR and DNA extraction procedures. If such selectivity did occur
during our procedures. it appeared to be treatment-dependent, because the
DGGE blots produced distinctly different probe hybridization results. The
consistent differences between treatments can be used to interpret how
environmental factors have influenced M05 in these soils, even if possible
technical bias precludes use of the results to describe actual in sifu populations.

. In contrast to the DGGE patterns, cloning and probe hybridization results
were not as reproducible from 1994 to 1995. This was due to the high
incidence of nonspecific (non-AAO) PCR amplification products in the 1995
clone libraries, which was probably due to two technical reasons. First, the
reverse primers used in the initial PCR reactions on soil DNA extacts were
slightly different in the two years. In 1995, we used the original B-AMOr primer
of McCaig et al. (1994). The previous year we used a modified reverse primer
from a 168 rDNA sequence starting three bases upstream from the original
primer. This modification had been made to reduce nonspecific PCR
amplification, specifically from DNA of Comamonas testosteroni. The use of the
original reverse primer in 1995 may explain the greater incidence of nonspecific
PCR amplification products observed. A second possible reason for higher
nonspecific amplification was that the 1995 soil DNA extracts were less pure
than 1994 extracts, because they had lower 0.0.250/230 and O.D.250/230
ratios. DNA quality in 1995 may thus have affected the specificity of the Taq
DNA polymerase in the PCR reactions. Nevertheless, the 1994 and 1995 clone

106
libraries were similar in that both exhibited higher frequencies of deep-
branched sequences in libraries from never-tilled soils.

These findings from 168 rDNA analysis are substantiated by results we
have obtained using cultural methods to estimate AAO p0pulation sizes in these
soils (Bruns et al., 1996). Most-probable-number (MPN) counts of M05 were
consistently lower by a factor of ten in never-tilled soils (104 AAOs per 9) than in
cultivated soils (105 M05 per 9). To characterize AAO population structure, we '
obtained simultaneous MPN counts for AAOs at various NH4-N concentrations,
an approach first used by Suwa et al. (1994). M03 in never-tilled soils had 60-
fold higher recovery in MPN medium containing 10 ppm NH4-N than in medium
containing 2000 ppm NH4-N. Recovery of M05 in cultivated soils was only
twofold higher at 10 ppm NH4-N than at 2000 ppm NH4-N (Bruns et al., 1996).
Greater recovery of MOS at the lower NH4-N concentration suggests that a
higher proportion of M08 in never-tilled soils do not grow out in NH4-N
concentrations above 10 ppm (Suwa et al., 1994). This is consistent with the
apparent predominance in never-tilled soils of 168 rDNA sequences
representing less easily cultured AAOs (Figure 4.3).

Consistently observed differences in the properties of cultivated and
never-tilled soils (Table 4.1) suggest that these soils provided distinctly different
habitats for the development of AAO populations. An important environmental
factor affecting AAO numbers and activity is their supply of inorganic N in the
form of NH4-N (Robertson,1982). Inputs of inorganic N were higher for
cultivated soils than for never-tilled and successional soils. Cultivated soils at
the LTER site received 84 kg fertilizer N per ha when the plots were cropped in
corn and 56 kg N per ha when the plots were in wheat (Robertson et al., 1996).
The method of fertilizer application used at LTER (broadcasting of ammonium

nitrate) would result in higher concentrations of NH4-N being localized at the

107

surface of the soil, due to rapid binding of NH4+ to soil colloids. Subsequent
incorporation of crop residues during moldboard plowing, as well as soil mixing
by disking and cultivating, would produce a more spatially homogeneous
distribution of inorganic N. The lower diversity of AAO 16$ rDNA sequences
from cultivated soils suggests that fertilization and disturbance have increased
selection for more homogeneous populations of AAOs. AAO populations in
never-tilled and successional soils may thus be more diversified as a result of
spatial heterogeneity in substrate availability.

, Another factor that affects the supply of inorganic N to AAO populations is
NH4-N uptake by heterotrophic microorganisms. Never-tilled soils have
significantly higher biomass C and bacteria per g of soil than cultivated soils
(Table 4.1). Thus, AAOs comprise a much smaller proportion of the soil
bacterial community in never-tilled soils, where they are outnumbered by
bacterial heterotrophs 364,000:1 (Table 4.1). AAOs are subject to less
competition for NH4-N by bacterial heterotrophs in cultivated soils, where they
are outnumbered 21,000:1. Lower fungal biomass in cultivated soils (D. Harris,
personal communication) would also result in less fungal competition for NH4-
N. Higher inorganic N inputs and less heterotrophic competition could thus
increase the supply of NH4-N to M05 in cultivated soils. These factors could
select for AAOs that favor comparatively higher NH4-N concentrations.

The cultural and molecular evidence that AAO populations were different
in these soils is supported by soil data indicating differences in nitrification
activity (Tables 4.1 and 4.7). Never-tilled soils in 1994 were characterized by a
higher mean concentration of NH4-N (10.5 pg per 9 soil) than cultivated soils,
which had a mean concentration of 2.6 ug NH4-N per 9 soil (Table 4.7). Never-
tilled soils had a mean N03-N concentration (0.5 pg per 9 soil) that was

significantly lower than the mean concentration for cultivated soil (3.9 ug NOa-N

108

TABLE 4.7. Indicators of the effects autotrophic ammonia oxidizers may have
on nitrogen cycle parameters, measured in two replicate plots of cultivated,

successional, and never-tilled treatments (1994 LTER data).1

 

 

Measurement Cultivated soil Successional Never-tilled
soil soil

Mean NOa-N concentration 3.9 (12.49) 0.4 ($0.21) 0.5 (10.24)

(#9 per 9 50")

Mean NH4-N concentration 2.6 (10.98) 3.8 (10.60) 10.5 (32.49)

(#9 per 9 80")

Net nitrification in field 0.17 ($0.10) 0.03 (10.5) 0.19 (10.15)

incubations (ug/g/day)2

Net nitrification in lab 0.35 0.18 0.29

incubations (ug/g/day)2

Net percentage of NH4-N 68% 30% 21%

converted to NOa-N during
field incubations3

 

1 Measurements taken monthly from May through November, 1994. Means are
from measurements for two replicate plots, with n=14. Standard deviations for
the means are given in parentheses (KBS LTER Site Data Catalog, KBS LTER
home page on World Wide Web, http:/kbs.msu.edu/lter/.

2 Field and lab samples were incubated for 21 days. For field incubations,
n=10. For lab incubations, n=2 (D. Harris, personal communication).

3 Percentage based on the net increase in NOa-N after 21 days divided by the

sum of the initial NH4-N pool plus the amount of inorganic N mineralized during
incubation. Percentages were obtained from inorganic N measurements before

and after incubation (KBS LTER Site Data Catalog, KBS LTER home page on

World Wide Web).

109
per 9 soil). Available data on net nitrification rates during field and laboratory
incubations of cultivated and never-tilled soils, however, did not show
significant differences (Table 4.7). Differences were observed, however, in the
percentages of total NH4-N pools (initial NH4-N plus NH4-N mineralized during
incubation) that were converted to NOa-N during the 21-day incubation period
(Table 4.7). Nitrifier populations in cultivated soils converted a larger
percentage (68%) of the NH4-N pool to NOa-N, whereas populations in never-
tilled and successional soils converted 21% and 30% of the NH4-N to NOa-N,
respectively. Nitrifier populations in the cultivated soils thus appeared to have
greater capacity to oxidize available NH4-N. This observation was consistent
with a predominance in cultivated soils of Nitrosospira spp. that were more
readily cultured in laboratory media. It was also consistent with the fact that
AAOs in cultivated soils gave higher MPN counts than AAOs in never-tilled soils
at 100 and 2000 ppm NH4-N.

MPN counts of AAO populations in never-tilled soils remained tenfold
lower than counts in cultivated soils, despite the fact that NH4-N levels in never-
tilled soils remained at around 10 ppm throughout two growing seasons (Bruns
at al., manuscript in preparation). One explanation for the apparent lack of
NH4-N consumption by these AAO populations is that the NH4-N was not
physically available in these soils. Never-tilled soils contained at least 1.5 times
more organic matter than cultivated soils (I' able 4.1). Biochemical interactions
between NH4+ and organic matter (Nommik and Vahtras, 1982) in the never-
tilled soils could have made the NH4-N unavailable for nitrification by AAOs (or
immobilization by heterotrophs.) Lack of disturbance in never-tilled soils could
also contribute to lower overall nitrification because of spatial discontinuities

between sites of bound NH4-N and locations of nitrifiers in soils.

1 10

Another explanation for low AAO population size in the presence of NH4-
N could be that nitrifier populations in never-tilled soils had lower metabolic
capacities to oxidize available NH4-N than nitrifiers in cultivated soils. Belser
and Schmidt (1980) and Suwa et al. (1994) showed that different genera and
strains of MOS exhibited varying Km and Vmax values for ammonia oxidation.
Thus, AAO populations in never-tilled soils could be dominated by strains with
relatively low Km and Vmax values for ammonia oxidation. AAOs with higher
Km values, on the other hand, would be more likely to be selected in cultivated
soils that receive fertilizer N.

Finally, another cause for low AAO population size in the presence of
NH4-N could be inhibition of MOS by allelopathic chemicals from plants (e.g.,
tannins). This explanation was put forth by Rice and Pancholy (1972) in studies
on nitrification in soils from grasslands and climax ecosystems. However, other
studies have not detected phytochemical inhibition of nitrification (Barford and
Lajtha, 1992), so that this explanation remains controversial.

In conclusion, this study presents strong evidence that AAO populations
in fertilized, disturbed soils were less diverse than AAOs in unfertilized,
undisturbed soils. Our results also indicated that N-fertilization resulted in
distinct differences in the kinds of Nitrosospira sequences retrieved from soil
communities. In addition, we present preliminary evidence that the molecular
analysis results in this study were temporally and spatially reproducible over an
area of 1 km2. Further testing of greater numbers of replicate soils sampled
more frequently over longer time periods can confirm the reproducibility and
reliability of these techniques in analyzing in situ AAO populations. Microbial
ecologists have long suspected that different habitats harbor different microbial
communities. This hypothesis about microbial diversity, however, could not be

tested due to the inadequacies of cultural techniques. Our study shows that

111
molecular techniques now available can be applied to the study of in situ soil
microbial diversity. This should facilitate integration of soil microbial diversity

into the development and testing of general microbial ecology concepts.

REFERENCES
Barford, C. and K. Lajtha. 1992. Nitrification and nitrate reductase activity along

a secondary successional gradient. Plant Soil 145:1-10.

Belser, L W. and E. L Schmidt. 1980. Growth and oxidation kinetics of three
genera of ammonia oxidizing nitrifiers. FEMS Microbiol. Lett. 7:213-216.

Bruns, M. A., M. R. Fries, J. M. Tiedje, and E. A. Paul. Evaluation of
Nitrosomonas europaea functional gene probes in hybridizations with
Nitrosospira DNA, manuscript in preparation.

Bruns, M. A., and E. A. Paul. Cultural methods for autotrophic ammonia-
oxidizing bacteria in soils, manuscript in preparation.

Donaldson, J.M., and GS. Henderson. 1989. A dilute medium to determine
population size of ammonia oxidizers in forest soils. Soil Sci. Soc. Am. J.
53:1608-1611.

Felsenstein, J. 1993. PHYLIP: Phylogeny Inference Package. J. Felsenstein,
Seattle, Wash.

Giovannoni, S. J., T. B. Britschgi, C. L. Moyer, and K. G. Field. 1990. Genetic
diversity in Sargasso Sea bacterioplankton. Nature 345:60-63.

Head, I. M., W. D. Hiorns, T. M. Embley, A. J. McCarthy, and J. R. Saunders.
1993. The phylogeny of autotrophic ammonia-oxidizing bacteria as
determined by analysis of 168 ribosomal RNA gene sequences. J. Gen.
Microbiol. 139:1 147-1 153.

Hiorns, W. D., R. C. Hastings, I. M. Head, A. J. McCarthy, J. R. Saunders, R. W.
Pickup, and G. H. Hall. 1995. Amplification of 168 ribosomal RNA genes
of autotrophic ammonia-oxidising bacteria. J. Gen. Microbiol. 141:2793-
2800.

Jukes, T. H., and C. R. Cantor. 1969. Evolution of protein molecules. In H. N.
Munro (ed.), Mammalian protein metabolism. Academic Press, New York,
NY. pp. 21-132.

Kowalchuk, G., J. R. Stephen, J. l. Prosser, T. M. Embley, and J. Woldendorp.
Denaturing gradient gel electrophoresis of specifically amplified 16S-

112

rDNA in the population analysis of ammonia-oxidising bacteria of the B-
Proteobacteria, manuscript in preparation.

Maidak, B. L, N. Larsen, N., J. McCaughey, R. Overbeek, G. J. Olsen, K. Fogel,
J. Blandy, and C. R. Woese. 1994. The Ribosomal Database Project.
Nucleic Acid Res. 22:3485-3487.

McCaig, A. E., T. M. Embley, and J. I. Prosser. 1994. Molecular analysis of
enrichment cultures of marine ammonia oxidisers. FEMS Microbiology
Lett. 120:363-368.

McTavish H., J. A. Fuchs, and A. B. Hooper. 1993. Sequence of the gene coding
for ammonia monooxygenase in Nitrosomonas europaea. J. Bacteriol.
1 75:2436-2444.

Manz, W., R. Amann, W. Ludwig, M. Wagner, and K.-H. Schleifer. 1992.
Phylogenetic oligodeoxynucleotide probes for the major subclasses of
Proteobacteria: problems and solutions. System. Appl. Microbiol. 15:593-
600.

Nommik, H. and K. Vahtras. 1982. Retention and vixation of ammonium and
ammonia in soils. In Nitrogen in agricultural soils, F. J. Stevenson (ed.),
American Society of Agronomy, Madison, Wisc., pp. 123-171.

Norton, J. M. and M. G. Klotz. 1995. Sequence of ammonia monooxygenase
subunit a in Nitrosospira sp. NpAV. Gene 163:159-160.

Norton, J. M., J. M. Low, and M. G. Klotz. 1996. The gene encoding ammonia
monooxygenase subunit-A exists in nearly identical copies in
Nitrosospira sp. NpAV. FEMS Microbiology Letters, in press.

Rice, E. L., and S. K. Pancholy. 1972. Inhibition of nitrification in climax
ecosystems. American J. of Botany, 59:1033-1040.

Robertson, G. P. 1982. Factors regulating nitrification in primary and secondary
succession. Ecology 63:1561-1573.

Robertson, G. P., J. R. Crum, and B. G. Ellis. 1993. The spatial variability of soil
resources following long-term disturbance. Oecologia. 96:451-456.

Robertson, G. P., H. P. Collins, S. H. Gage, K. L. Gross, S. J. Halstead, R. H.
Harwood, K. M. Klingensmith, M. J. Klug, and E. A. Paul. 1996. Long-term
research in agricultural ecology: objectives and establishment of a site in
the'U. S. Midwest. Agric. Ecosyst. Env., in press.

Saitou, N., and M. Nei. 1987. The neighbor joining method: a method for
constructing phylogenetic trees. Mol. Biol. Evol. 4:406-425.

113

Sayavedra-Soto, L. A, N. G. Hommes, and D. J. Arp. 1994. Characterization of
the gene encoding hydroxylamine oxidoreductase in Nitrosomonas
europaea. J. Bacteriol.

Schmidt, E. L. and L W. Belser. 1982. Nitrifying bacteria. In A. L. Page. R. H.
Miller, and D. R. Keeney (eds.) Methods of soil analysis, part 2. Chemical
and microbiological properties. American Society of Agronomy, Madison,
WI. 139. 1027-1041.

Schmidt, T. M., E. F. DeLong, and N. R. Pace. 1991. Analysis of a marine
picoplankton community by 168 rRNA gene cloning and sequencing. J.
Bacteriol. 173:4371-4378.

Stahl. D. A., B. Flasher, H. R. Mansfield, and L Montgomery. 1988. Use of
phylogenetically based hybridization probes for studies of ruminal
microbial ecology. Appl. Environ. Microbiol. 54:1079-1084.

Stephen, J. R., A E. McCaig, Z. Smith, J. I. Prosser, and T. M. Embley. 1996a.

Molecular diversity of soil and marine 16$ rDNA sequences related to B-
subgroup ammonia oxidising bacteria, manuscript submitted to App.
Environ. Microbiol.

Stephen, J. R., G. Kowalchuk, M. A. Bruns, J. I. Prosser, T. M. Embley, and J.

Woldendorp. 1996b. Analysis of B—subgroup chemolithotrophic ammonia
oxidiser populations by DGGE, membrane transfer and hierarchical 168
rDNA oligonucleotide probing, manuscript in preparation.

Suzuki, M. T., and S. J. Giovannoni. 1996. Bias caused by template annealing
in the amplification of mixtures of 168 rRNA genes by PCR. Appl.
Environ. Microbiol. 622625-630.

Swofford, D. L. 1993. PAUP: Phylogenetic analysis using parsimony. Illinois
Natural History Survey, Champagne-Urbana, lll., USA.

Teske, A., E. Alm, J. M. Regan, S. Toze, B. E. Rittmann, and D. A. Stahl. 1994.
Evolutionary relationships among ammonia- and nitrite-oxidizing
bacteria. J. Bacteriol. 176:6623-6630.

Voytek, M. A, and B. B. Ward. 1995. Detection of ammonium oxidizing bacteria
in the beta-subclass of the class Proteobacteria in aquatic samples with
the PCR. Appl. Environ. Microbiol. 61:1444-1450.

Woese. C. R., W. G. Weisburg, B. J. Paster, C. M. Hahn, R. S. Tanner, N. R.
Krieg, H.-P. Koops, H. Harms, and E. Stackebrandt. 1984. The
phylogeny of the purple bacteria: the beta subdivision. System. Appl.
Microbiol. 5:327-336.

114

Zhou, J., M. A. Bruns, and J. M. Tiedje. 1996. DNA recovery from soils of diverse
composition. Appl. Env. Microbiol. 62:316-322.

APPENDIX A
DNA RECOVERY FROM SOILS OF DIVERSE COMPOSITION

115

APPLIED AND ENVIRONMENTAL MICROBIOLOGY. Feb. 1996. p. 316—322

116

0099-22-10!96.‘SO4.00*0
Copyright C) 1996. American Society for Microbiology

.-

DNA Recovery from Soils of Diverse Composition

JIZHONG ZHOU, MARY ANN BRUNS. AND JAMES M. TIEDJE‘

Center for Microbial Ecology and Department of Crop and Soil Sciences,
Michigan State University, East Lansing, Michigan 48824-1325

Received 9 May 1995/Accepted 7 November 1995

A simple, rapid method for bacterial lysis and direct extraction of DNA from soils with minimal shearing was
developed to address the risk of chimera formation from small template DNA during subsequent PCR. The
method was based on lysis with a high-salt extraction buffer (1.5 M NaCl) and extended heating (2 to 3 h) of
the soil suspension in the presence of sodium dodecyl sulfate (SDS), bexadecyltrimethylammonium bromide,
and proteinase K. The extraction method required 6 h and was tested on eight soils dilering in organic carbon,
clay content, and pH, including ones from which DNA extraction is dil‘icult. The DNA fragment size in crude
extracts from all soils was >23 kb. Preliminary trials indicated that DNA recovery from two soils seeded with
gram-negative bacteria was 92 to 99%. When the method was tested on all eight unseeded soils, microscopic
examination of indigenous bacteria in soil pellets before and after extraction showed variable cell lysis
eliciency (26 to 92%). Crude DNA yields from the eight soils ranged from 2.5 to 26.9 pg of DNA g“', and these
were positively correlated with the organic carbon content in the soil (r = 0.73). DNA yields from gramopositive
bacteria from pure cultures were two to six times higher when the high-saIt-SDS—heat method was combined
with mortar-and-pestle grinding and freeze-thawing, and most DNA recovered was of high molecular weight.
Four methods for purifying crude DNA were also evaluated for percent recovery, fragment size, speed, enzyme
restriction, PCR amplification, and DNA-DNA hybridization. In general, all methods produced DNA pure
enough for PCR amplification. Since soil type and microbial community characteristics will influence DNA
recovery, this study provides guidance for choosing appropriate extraction and purification methods on the

Vol. 62. No. 2

basis of experimental goals.

 

Isolation of bacterial nucleic acids from natural environ-
ments has become a useful tool to detect bacteria that cannot
be cultured (11. 27), to determine the fates of selected bacteria
or recombinant genes under natural conditions (10, 19), and to
reveal genotypic diversity and its change in microbial ecosys-
tems (22). Many workers have attempted to increase DNA
yields from soils by using severe physical treatments such as
mechanical bead beating and sonication to lyse indigenous
microbial cells. Such treatments can shear DNA to sizes of 5 to
10 kb or less (11, 14), and in at least one study, the average
fragment size was 100 to 500 bp (17). Such DNA may not be
suitable for community analysis based on Taq DNA PCR,
because of the risk of forming chimeric products with smaller
template DNA (12). Because microbial cells may remain
tightly bound to soil colloids. soils high in clay or organic
matter pose particular challenges to obtaining high yields of
high-molecular-weight DNA. Most DNA extraction methods
have been tested on a limited number of soil types, so that their
general applicability is unknown for comparative ecological
studies.

Extraction of DNA from soils always results in coextraction
of humic substances which interfere with DNA detection and
measurement. This contamination can inhibit Taq DNA poly-
merase in PCR (18, 25), interfere with restriction enzyme di-
gestion (15), and reduce transformation efficiency (21) and
DNA hybridization specificity (19). Since humic substances are
difficult to remove. DNA purification is a critical step following
direct extraction to obtain DNA of sufficient purity.

Objectives of this study were to evaluate and improve DNA

‘ Corresponding author. Mailing address: Center for Microbial
Ecology, Plant and Soil Science Building. Michigan State University,
East Lansing. MI 48824-1325. Fax: (517) 353-2917. Electronic mail
address: 21394jmt@msu.edu.

extraction and purification methods for speed and simplicity,
DNA yields. DNA fragment size, and applicability to a broader
variety of soils. We tested these methods on eight physically
and chemically distinct soils, including soils from which DNA is
difficult to extract and purify. We emphasized PCR amplifica-
tion in evaluating DNA purity because Taq polymerase is sen-
sitive to humic contamination and because PCR amplification
is a major use of extracted soil DNA.

MATERIALS AND METHODS

Soils. Eight soils were used to evaluate the efl‘iciency of DNA extraction and
purification procedures. Six of these soils had been selected from a global soil
collection (9) to represent a range of soil properties (Table 1). The other two
soils. Native Kellogg (NK) and Cultivated Kellogg (CK). were obtained from the
National Science Foundation Long-Term Ecological Research site at Kellogg
Biological Station near Kalamazoo. Mich. NK and CK soils were from the same
soil series (Kalamazoo sandy loam. typic hapludalf). but CK soil has been cul.
tivatcd for the last 40 years while NK soil has been undisturbed. All soils came
from regions having predominantly luvisolic soils. as described under the Food
and Agriculture Organization Soil Classmcation System (7). The six soils from
the global collection had been sampled between 5 and 30 cm in depth in 1993.
NK and CK sons were sampled between 0 and 15 cm in 1993 and 1994. All soils
were kept on ice or stored at 4°C until they were tested in the laboratory.

Soil moisture contents were determined by drying at 110°C for 48 h. Particle
size analyses were performed by a modified hydrometer method. in which the
clay content was determined after 8 h (4). Carbon and nitrogen contents were
determrned on oven-dried. ground samples in a Carlo Erba NA 1500 series 2
nitrogen/carbon analyzer (Fisons Instruments. Beverly. Mass). Soil pH was de-
terrnined in a slurry (5 parts distilled water. 1 part soil). Soil color was evaluated
by Visual examination in outdoor sunlight with Munsell color plates.

Bacterial strains and soil inoculation. Pseudomonas sp. strain 813 (S) was
used as the seed organism. Cells were grown to late exponential phase on M9
medium supplemented with trace minerals and 5 mM 3-chlorobenzoatc and
resuspended in 2 ml of extraction buffer (see below) before beIng Inoculated into
the soils. This cell suspension was mixed with sterilized soils. which were ob-
tained by autoclaving twrce at 121°C for 00 min. Seeded soils were kept at room
temperature for 30 min prior to DNA extraction.

Etect of CIA! and PVPP on humic contamination of crude extracts. Hexa-
decylmethylammonium bromide ( CT AB) and polyvinylpolypyrrolidone (PVPP)
have been used in previous studies to complex and remove contaminants from

VOL 62, 1996

"marvel-villi) alnitnatflv WI 900.4 ‘ovs .

 

uieol .(irp .(puas (aw) eglaitsnv 'ugpanaw

snouliqdoiops/utioutm nigpow

losteuo; oguiuax

[/7 HA 5'
{/9 HA S'L

I '6
1’")
I '9

U)

690

9011

(HA) tire.) ‘VIIIH 3310M
(41) onto ‘snpnuod soitr]

(>0) 'lPIW ‘fiottox POIWIIIO

tuntr]
ipuas

snolliqdoiops/utroueuoitpaw

|0SIAII| ogtuoItD
IOSIAII| Jituolqg)

80'0

.(putrs

0m) 'Itaiw ‘flortox mum

(mi) Btwtu Mom mm

(Lg) uamoqawxsus ‘qung
(AM) WMIPIBIIWS ‘OIIWIVM

snoqudoIcps/ulcuauoiipaw

tosytnj omuo potmpmtt snonpgoop/atuiodtuoy
1051*"! Drum poomwuu monpmp/mmdwai

Z/f.‘ )IA OI

(.61)

“1'0

Z/I.~ HA 0|

96'0

09'1

800

V9

III: “A 01

team; luaioq/aituaduroL

lost/mu otcuv
ImiAnt nutty

H." 81'

EH)

[/8 21A 01
EliI 8A 0|

91111 Us

9011

1:310; walth/Olmodwai

WSW" ’l‘IIV

O/Z 21A S'L

I")

98's

1911

 

unlllnul Zuilduitrs nos

:untxnt 110$

nonunion mljllu/I‘IIEWID

IIIS W's luidnms
=10 1%) my "m ‘0‘”

£20

at, tiniuoo
minnow

(alumna/onion '3ntt)
Iqoa mow

on 11"

N96

 

sompaoold uotiotuixo VNG ug posn suos yo squadord pun suogtnaol flutldiuirs 1 may;

117

DNA RECOVERY FROM SOILS 317

DNA (1. 10). Seeded NK soils were used to evaluate the cflect of CTAB and
PVPP in the extraction buffer on humic contamination in crude extracts. Soil (5
) was mixed vnth extraction buffer (see below) containing (i) no CI'AB no
PVPP. (ii) 1% CTAB. no PVPP: or (iii) no CTAB 2 g of PVPP. Soil suspensions
were. then processed by the extraction method described below. Spectrophoto-
me erit Wkly", andA osw redetennined to evaluate levels of
protein and humic acid impurities. respectively in the crude extracts 14.20).
S S—hascd D ccCl‘AB performed better in reduc-
Ing humic contamination. it was used In the bufier for sodium dodecyl sulfate
(SDSrbased DNA extraction Soilsarn les sof 5 g wercrn ixed with 13 5 ml of
NA extraction bufier OOmM Tris-HCI [pH 8 0.11100 mM sodium EDTA [pl-l
8. O] 100 mM sodium phosphate [pH 8 0]. 15 M NaCI 1% CTAB) and 1(1) pl of
proteinase K (10 mgtml) In Oakridge tubes by horilzontal shakingw at 225 rpm for
30 min rthe shalungtr e.atment 15m oI 20% SDS d
the 5am p1es were Incubatcdin 465°C water bath for 2 h with gentle end-over—end
inversions every 15m 20 min. supem ata ants were co ected eroe ntrifuga-
tion It 6.011) X g for 10 mitt at room temperature and transferred into SO-ml
J - L Jr A c I

I O
ofth extraction buffer and 0.5 m1 of 20" /! SDS. vortexing for 10 s. incubating at
65"C for 10 min. and centrifuging as before. Supernatants from the three cycles
ctionsw dnda mixedwi han equal volume of chloroform-
iosoamyl alcohol (24: 1 vol/vol). Theaq eous phasew rccoverbved centrifuga—
tion and precipitated wit ih 0 6 volume of isopropanolas at roo mtcmperature for l
h.11ie pellet ofcrude nucleic acids was tained byce centrifugat iona
g for 20 min al room temperature. washed with cold 70% ethanol. land resus-
pended in sterile deionized water. to give a finial volume of 500 al.
Cell lysis and LLCel Ivstseliciency was estimated for
2.11231: Iccolunts of soil smears (3) obtained
A extraction. 5 to 15

six of the eight soils by directrac micr
before and after the DNAex LoreBef D

of soilw ashlended In a Waring blender for 1 min in 150 to 190 ml of sterile.
filtered (0 2 jun) deionizedwa wa.Ier The ecoarse particles in the blended slurry
owed to settle for l min. and then a 10-ml subsamplew moved from
the upper portion of the surrvl to a 1.5—ml sterile tube. The
subsample was vortexed for 10 s. and 4- or I6-iiJ aliquots were removed for
smears. Each aliquo wasevesprud ninly rcle on a coated slid
(Cel- -Line Associates. Newfield NJ.) and quickly dried at 40°C. After DNA
extraction. smears werehp prep: arcd from the remaining soil pellet and the pooled
tatsn sampIcA subsample of soil pellet was reemcw ctd
determine ther0 moistur: content. and the reima He was blended with
appropriate amounts of water to prepare smears. Before DNA was precipitated
and quantified from each supemaianL a loo-til subsample of the supernat ant

ilu ted 1: 201m water to makes smears

aned smears were flooded with 10 id of a fluorescent staining solution con-
tainingm 2 mgo of DTAFs [5- (4. 6-dichlolrotn'azin- 2-yl) )arnino IIu orescein :Sigm
Chem loC ..St Loui .Mo. )1 per ml of bu let (0.85% NaCla. 50 mM
Nazi-IIi'O,I [pH 9.0)) (16). Hooded slides were held for 30 min covered
container to prevent drying. The slides were rinsed bv' miners on in fresh bufier
thr ree times In or 20 rn '
dried. Slides were stored in the dark at 4‘C for no longer than 48 h before
microscopic analysis

The were examined with a 63X objective on a leitz Orth oplan 2
and a Leitz 13 filter
excitation filter. RKP 510be am splitter. and LP 515w ss~uppre
device camera (Princeton Instruments Trenton.

epifiuorsesccncem
block (8? 450-190
sion filter) A charge-co upled
NJ.) was used ‘,' ‘ , , , w _
selected at random along two central transects (8). Images were transferred to a
Power Macintosh 7100166 via an S'I'13S detector/controller and GPIB interface
card (National Instruments. Austin Tex) for display by 11’ Lab Spectrum Image
analysts software] (S igna alAn Analytics Va.) Bacterial cells were
counted byVI culmination of images on the computer monitor. All counts
were obtainedsu by one investigator.

DNA

 

gran- tive bacteria. The gram-positive bacteria used
for comparing cell lys'n methods were IgrownN on tryptic soy agar at 37°C over-
right“

 

in terms of DNA yield and fragment size: (1) grinding freezing-thawing. and

SDS: (ili) Ireezing- thawing an nd SDS: and u(iidi) SDS. Cell pellc.tsd from purl: cul-

turcst .10 0I" ls‘ml ')were with am orrta dtcpecl the

of sterile sand and liquid nitrogen before addition of extraction bufier
—7(7"I‘

 

IndSD S.“
heating until they boiled briefly. a total of three times; then the DNA was
extracted byfo llowt‘ng a protocol similar to the one desert bed a.bove The DNA
yield was determined hotomerr'yI

nua

of crude extracts were evaluated. One-tenth to one- -hfth of the crude DNA

on
at

turd PCR reps rt resinj
ii) double minicolumn (eluate from the firstrcsl nunico puri-

_ lied furthei'l with fresh resin and passage through another minicolumn); (iii) gel

 

{image of the excised and melted gel hand through a Wizard minicolumn); and

118

318 ZHOU ET AL.

TABLE 2. Compiirison of DNA yield Lll‘ld purity of the crude DNA
from seeded NK sod Subjected to dillcreni treatments“

 

 

.7 DNA )IC‘tIS (Iigig .4 mini. A: .1143.
Tnntment [dry wt] of wil)" {MINE-w rhitti"
No PVPP. no CTAB 17.1 1' U.‘) I 17 z (1.02 (1.72 : l).l)3
CTAB. no PVPP 17 S t 12 1.35 : ()04 0.91 :1).l)3
PVPP. no CTAB 10.9: 1. 5 | 23 r 0.05 [1.88 I 0.03
Purc culture 1.8“ 157

 

" NK soil was sampled In 1‘19] and altllltd tit 4°C for Ii months.
" DNAyc 1d: (it man \.ilucs s1.i.ind id \IL‘VIIIIIUII) w: c dL‘ lerrtiiited
hyc lllluuromL-Iry The ldIIUS were calculnllctl from spectrtipholoiiictric measure»

,Ii=t:

(iv) 'gel plus centrifugal concentrator. (The crude curuct was subjected to gel
-ii in II II'III

 

I sis mil [1 )CD

 

[Epicc IiIrL- Technologies MtltllVH] Wis I by lolluwingl the rapid protocol of ”[110
Ictrniiulncturer Nucle L':iL‘Id\ wch lch .Islied and concentrated in .i CL'nI

i-fill [AmiL tin CL orp. BcVL‘ rlv. Mass | )W
clhe niiiticnlumn CIIpLiL'ilv w.is1imilcdto 1 ml III resin only .‘i fraction (1.5
to 1‘nlll) o! tltc‘ crutlc L'ttrnt't from 5 3L Lsil Liil could be purilicd .II I lime
liirgL-r-sculc geusl-pl column purification prL ICL gdiir susi: LLl In pii nrify ihc Lntirc
clriIdL- thracts lmmfi -g samples of the tight test so ||1\ Lm‘ procedure employed
Umlc CIiap ciiyL column “’1”! nlWimrd Minipreps .ims id I'll‘
rlilicIIIion rL.[ri1’ttimL' _;ij) ltecriiiw [116 to sin used in the muncoliimmcr iiitninLd
.i I‘ll-iim- absorbing sul’isluiiLL| sili. it iiti turtL-rcd With schimpliiittiinctric Ittcnsiire-
niL III (II DNA. cDNAw clutLd lrorri :hL resin twicc with rill) )LI iil hot
(71) (_) TI|\- EDTA Cliullcr 1U 1|1IC11|1.11I. rLle .IsL- til high :Itulcculnr-WCIgltt DNA
antification AILl sn .ill -sL.ilL
Iluortimetry yiih :i TK I()( rIiiliorLrnLiLr llloclcr
Frantitc .:.Cilif) by following the L'\'1'
nt I.iniiliit‘turL-r Tlt IL- lluo iront Lier \1V is cnlilir ItLd with he rring sperm Dm.N rl\ BOL lir
iiIgLr M Iitnl lm.1l’1t.1|.ll|.lp01|\. 1nd) Tltc DN A yL-i les were estimated on the
brisis of “1.10:1“ Iltrcg rL-pli‘c‘ IiIL dc m.rn.itiiins
DNA

‘rilL ,nuiliLuuL i. '

 

 

 

InthsiIii: s til crime I) In :igzi riisc "L | hunds in van riL-d l’olzimitl pliiic
Ciiitlg .ind filllll DNA L-tImLIx \vcrg \LIDJCCIULI to electriipiiiirlgscis in lris- L'L~

EDT \ (TAE) butler containing (Li ug UI ctltidiiirit limn epcr nil in (1.7%
agarose gel containing DNA standards til \ Ito litg ()1 ltlmtbdii phage DNA
(1% K1! tcring Lr :Miml icli in 1). GLI pholtigfllplfl itLrL sc. iniiL- |wrth II Heivlctt~l‘.icl..ird
tile 11: i ner ganr: iting digitized ini.i_gLs [ILII \v-rL'LthIVZL‘IIWIllI1P1..‘lh
Spectrum stiflwnr re A strtn .i iril cum 1 N.-\ (tittccntmtion ill) to 51) ng ()1
ms prL Lp.iretl Iiir I“LuiLh gel and

 

Cdi Igctts wt: IL performed With J
DNA .i "(14 U o.nli endULllllClCJSC (BlllllHI. I)ml tLiiRl. 12w

.Yliull in Ill ul iiIlItL‘ .ipproprmic iiLlI r .I\ pri VlLlL d by ilIL miniilziLiurLr All

IlILIlhitllllnltJI1_lU1DI1 II‘L' DN\ {I.igniL-ntstyLrL rLsiiIvie I in i 1'

Tu ev I1|lIIIL DNA llVI‘rldlZMllllll’l l) 5 ugt

(\lltncn

transferred to :i Gent-St LrL‘t-rt Plus mentbrunc (Dupont. H
hr idinrron hybriilu ntiiin .inLl \v.inisli rigs WL rL- mirrictl 1‘ iii .Is‘ dLsL‘rilrL-Ll preiitiusly
(2‘ l.) A IV-LR amplified hlS-bp lrzignicnl (1'11“: LILD gene was mu] m the pm

' RESULTS

Soil properties. The physical and chemical properties of the
eight soils used in the DNA extraction study were quite differ-
ent (Table l). Soils were classed as loanis sandy lonms or
sandy clay lozims. with clay contents ranging fromS

The WV soil had the highest organic C content. which 715
reflected in its dark color (chroma = 0). and the highest N
content. The ME soil had the lowest organic C and N contents.
the rcddest color, and the lowest moisture content. The pH of
the soils ranged from 4,810 9.1.

Elect of CTAB or PVPP on DNA extraction from NK soil.
When seeded NK soil was treated with different extraction
bullcrs. no difference in DNA yield was observed among [retit-
ments with or without CTAB. Significant dillerenccs in DNA
yield did occur among treatments with or without PVPP (Table

2.) The crude DNA solutions front PVPP or CTAB treatments
were lighter in color and had higher AzNJAW, and Amy/1...”

APPL. ENVIRON. MICROBIOL

ratios than did the solutions from treatments without PVPP or
CTAB (Table 2). Both CTAB and PVPP can effectively rL-
move humic mLIICIIillh. but unlike PVPP CTAB resulted in no
DNA l 055. CTAB Lind PVPP did not completely remove humic
compounds since the/l Aim/An" rind/1 .m/Aw, ratios tor Crude

oils were significantly lower than the ratios tor
DNA solutions from pure Liilturcs (Table _).

DNA extraction ellicicncy WilS determined by comparing the
total crude DNA obtained from it known cell density of sterile
seeded s0il samples With the DNA extracted from the pure
culture, Crude DNA recoveries from seeded NK and CK soils
were 92 to 9‘1’1.

No significant differences in DNA yields were observed
when the crude DNA was allowed to precipitate in isopropnnol
for 1 h or overnight, either at room temperature or .it —2l)°C
Repeated extractions of the soil pellets were beneficial. .since
small amounts of DNA were still recovered after the second
and even the third wash. depend rig on the sotl (data not
shown). In the optimized method used subsequently. soil pel-
lcts were extracted three times. Most of the soil DNA frug-
mcnts were larger than 23 kb and similar in size to DNA
isolated from pure cultures (Fig. 1). These results suggest that
the extraction protocol did not cause severe shearing of DNA.

Evaluation of DNA extraction and cell lysis on more chiil-
Ienglng soils. Crude DNA was extracted front eight unseeded
soils by the SDS-DHSL‘LI method with CTAB. and mean yields
ranged from 2.5 to 20.9 ug of DNA per g (dry weight) of soil
(Table 3). The WV soil Iizid the highest DNA yield. and the
ME soil had the lowest yield. Significant correlation was
observed between crude DNA yield and .soil organic C content

I = 0.73. P = (1.01).

Lysis ellicicncies of the DNA extraction procedure for six
soils were estimated by microscopic examination of soil smears
before extraction and of soil pellets rind pooled supernatnnts
utter extraction, No cells were found in any of the 1:20 dilu~
tions of the pooled supernatants. but cells were found in the
residual pellets (Table 4). Postextrnction counts of WV. RU.

. T soils were significantly lower than preextrnction
counts, indicating high lysis ctlicrcncies (67 to 92%). Two soils.
VH and ME. appeared to show poor lysis by this method.
Significant correlation was observed between cell lysis efli-

kb

to»

 

FIG Agarose gel electrophoresis of total DNA ctlriit‘tcil Irunt NR \011 by
diflcienl treatments ‘irics 1 Hiiitil ll-ciit li. ILtL-riiipli. .igL 1.im|\d.initili.cul.Ir ‘llk'
marker r(|).t g)2 piirc-culiiiri- DN.\ \liu ini Pt w iu/iiiiriiiiiii sl" NIT ill" I "C"
soil. 4. “fr”; sot Iplu s.Bl‘l s, nnnsicrilL \‘1l1 DNA Lttr. Med with out I’V I’l‘ ind
CIA B'i nsiL rie \011 DNA L-xirzicIL-d \Villl P\ W, 7, nunsterilc soil DNA
extracted with Cl AB

119

VOL 62, 1996 DNA RECOVERY FROM SOILS 319

TABLE 3. Crude DNA yields from eight soils and percentages of crude DNA recovered in the large-scale purification method involving gel
electrophoresis and passage through one Megacolumn with 10 ml of Wizard Minipreps plasmid purification resin

 

 

Soil Crude DNA yield‘l Final DNA yield‘l Crude yield Final yield 9% of crude
(ugx’g [dry Wt] of sail) (pg/g [dry wt] of soil) CV (9%)" CV (92)” DNA recovered
WV 26.9 I 6.5 20.1 I 5.0 24 2.5 75
BT 12.5 2 5.1 3.4 I 0.9 41 28 27
RU 13. I 2.3 8.4 t 2.8 17 34 ()1
NK 2|.6 : 5.1 12.0 I 1.3 24 11 56
CK 4.9 I 1.1 3.9 I 1.2 22 31 80
LP 5.7 1': 1.1 3.1 I 1.4 20 44 54
VH 3.0 I 1.1 2.3 i 0.5 35 22 77
ME 2.5 I 0.6 2.0 it 0.6 22 31 80

 

" DNA yields were estimated by gel staining. Yields are mean values (: 1 standard deviation) with n = 5. except for NK and CK soils. for which n = 6.

” 7V. coclficrent of variation.

ciency and clay content of the soils (r = —0.67; P = 0.01). Lysis
was also evaluated by comparing the maximum and minimum
expected DNA yields based on literature estimates of cellular
DNA content for soil bacteria (2, 23). DNA yields for soils
WV, RU, NK, CK, and ME were within the ranges of expected
DNA yields, but observed values for BT, LP, and VH soils
were below the minimum expected DNA yields (Table 4). By
this method of evaluation, the ME soil gave the expected
amount of DNA.

Comparison of dilerent methods for lysing gram-positive
bacteria. Since the SDS-based method may not have lysed
some gram-positive bacteria, we evaluated two more physically
severe cell lysis methods on pellets from pure cultures of gram-
positive bacteria. The DNA yield was two to six times higher
for most of the bacteria examined by the grinding-freezing-
thawing-SDS method than by the freezing-thawing-SDS and
SDS methods (Table 5). While a very small portion of the
DNA was sheared by the grinding-freezing-thawing-SDS
method, most of the DNA had a high molecular weight and a
similar size to the DNA from the freezing-thawing-SDS and
SDS methods.

Comparison of DNA purification methods with crude DNA
from NK soil. To evaluate DNA purity for enzyme digestion.
PCR amplification, and DNA hybridization, four purification
methods were compared by using portions of the crude extracts
from seeded and unseeded NK soil. Because of the small

capacity of the minicolumn used in these methods, only 1/10 of
the crude DNA extract from 5 g of soil was purified at a time.
All four purification methods resulted in complete removal of
the brown color from crude DNA solutions. DNA recovery
varied with diiferent purification methods. Higher recovery
was obtained with gel-plus-concentrator purification than with
column methods (Table 6). However, recovery by gel electro-
phoresis was more variable and depended on the size distribu-
tion of DNA fragments in crude extracts. In addition, loss of
DNA was greater for the first minicolumn purification (~20%)
than for the second minicolumn purification (~5 to 6%) (Ta-
ble 6). This suggests that humic materials in crude extracts
might interfere with DNA binding to the resin.

Restriction endonuclease digestion was possible only with
purified DNA. While all enzymes cut the DNA purified by the
gel-plus-minicolumn method, DNA resulting from single-mini-
column purification was only partially digested by most of the
enzymes (Table 6). DNA quality could be improved by a sec-
ond minicolumn purification, since the eluted DNA was di-
gested by most of the enzymes. While all of the purified DNA
samples were completely digested by BamHl, most methods
resulted in DNA that was only partially digested by HindlII
(Table 6).

Amplification of the 16S rRNA genes was successful when
DNAs purified by all tested methods were used as templates
(Fig. 2). No PCR products were observed with DNA from

TABLE 4. Direct counts and DNA yields for individual samples of eight soils

 

 

Soil Preextraction count/g Postextraction countjg % Lysis Expected DNA yield‘ Crude DNA yield"
(dry wt) of soil” (dry wt) of soil efiiciency" (ug'g [dry wt] of soil) (ugtg [dry wt] of sorl)
wv (9.7 : 0.6) x 10° (2.0 2 02)“ x 10" 79 15-50 33.5
BT (7.3 t 0.4) X 109 (5.7 1: 2.1)‘ X 10" 92 11-39 8.7
RU (6.9 1 1.1) X 10‘J (1.4 t 0.1)” X 10" 80 9-40 15.5
NK '. z . x 0° ' 'D 7— 19.6
CK 8.5 t 5.9; x 10" 2‘3 lilo 433; ..9
LP (4.6 1' 0.5) X 10” (1.5 t 0.2)‘ X 100 67 7—26 4.6
VB (3.5 : 0.3) X 10" (2.6 I 0.4) X 10" 26 5-19 2.0
ME (1.3 t 0.1) X 109 (4.2 t 0.6) X 10° —" 2-7 2.3

 

” Mean count : standard deviation (two smears per sample).

” Determined from one sample of each soil and calculated from mean counts: [100 — (postextraction countr’preextraction count)] X 100.

' Range of expected DNA yield obtained by multiplying lower and upper limits of precxtraction counts (mean counts : 1 standard deviation) by low and high
literature values reported for cellular DNA content of soil bacteria (1.6 t‘g cell ' '. as reported by Bakken and Olsen [2]. and 5 fg cell " ‘. as suggested by Torsvik and

Goksoyr [23]).

" DNA yield from the supernatant associated with pellet used for the postextraction direct count. hie DNA yield was estimated by gel staining.
‘ Significantly diflerent from the preextraction count at the 5‘7: level.
1 ND. not determined.
" Not determined. since postextraction counts were higher than preextraction counts. Higher counts in ME soil following DNA extraction could have been due to
cell masking by soil particles in preextraction ME smears. because they contained more soil (1.1 mg em'z) than did preextraction smears from other souls (0.1-t to 0.7
mg cm 2). Bloem et al. (3) have recently recommended a maximum soil density for direct counts on loam soil of 0.8 mg cm‘ I to minimize cell masking by sorl particles.

120

 

 

 

320 ZHOU ET AL. Am. ENVIRON. MICROBIOL.
TABLE 5. Comparison of DNA yield from gram-positive bacteria" by different lysis methods
DNA yIL‘ld (tLEUIAi) from
Treatment

.4 plain/omits B ruh/r/u C. rr'mrlt' M lit/em R r-n'I/impulr's EuJQ Ben-ZR S lil‘lfllllll
Grinditig-freezmg-thawtng-SDS 0.47 U 33 0.40 0.22 0.23 0,19 0.26 0.37
Freczmgthttwmg S 0.13 0.1‘) 0.46 0.12 ND” ND ND 0.08
SDS 0.11 0.16 0.43 0.12 0.10 0.07 0.115 0.00

 

“The grilm- positive bacteria are inlrrobuuur gInhi/mniir. Baal/us rirliIr/ir ('it/itte/Irltlentrrti rt vmh

strain Hm). Rim/1c: c'omtx sp.s tratn Be n- 18.4 rid YlltplwtttcnrInn/mu The UN
"N e.d

.not detc rnitn

crude extracts. No amplification products were detected when
target DNA sequences were not present in the reaction mix-
tures. A second set of primers (for the c/cD genes) was used to
further test for PCR amplification in parallel seeded soils. The
DNA templates purified by all methods produced the expected
product (data not shown).

DNA extracted from seeded NK soil was used to determine
whether the DNA purified by each of the four methods was
pure enough for Southern hybridization. Crude and purified
DNA extracts and DNA obtained from pure cultures of the
same strain were hybridized with the cch gene probe. Signal
intensitiesw were very similar between pure- c-ulture DNAe
tracts and purified soil DNA from each ot the four methods
(Fig. 3). Weak hybridization was observed for the crude DNA.
The DNA purified by all tested methods was pure enough for
Southern hybridization.

Evaluation of purification methods on crude DNA extracts
from more challenging soils. Small-scale purification methods
were also evaluated with crude DNA extracts from the six
global soils. Only the double-minicolumn and gel-plus-minicol-
umn methods resulted in complete removal of the dark color
from all six crude DNA solutions. The gel- plus concentrator
method did not completely remove the dark color
WV and RU extracts. probably because of the high organic C
contents of these soils. Crude extracts from WV. RU, and NK
soils till contained greater amounts of high molecular-weight
humic acids as observed during electroplioretic separation of
D A rom humic contaminants After e ectrophoresis excised
DNA bands could have contained such contaminants which
cannot be washed through concentrator filters with lower nio-
lecular weight cut tsoff

PCR amplification of the 168 rRNA genes was successful
with DNA purified by the gel- plus- -minicolumn an nd g-el p-lus-
concentrator methods. DNA urified by the double- minicol-

umn ethod was amplified in only four of the six soils (WV
and RU soils produced no PCR amplification), indicating that

TABLE 6. DNA recovery for different purification methods
an enzyme digestion ofthL ' DNA Asa mplcs
pur rifie d from unseeded NKs soil

 

Digestion" by:

 

Purification Recovery"

'"°”'°" 1%) 131:"th Drul Ele EcnRV Hmdlll .Ylinl
Crude extracts 100 — — _ _
Column 80 (t : 2.1 + z z : ~ :
Column + column 74.5 : 3.8 + t + + 3 4.
Gel + ccntricon 91.4 t 7.5 + + : 4- : *
Gel + column 84.1: 7| + + + + + +

 

" Percentage of DNA recov vcrcd as measured by UV absorption compared
With DNA In crude extract _ l stttntlard deiiatto
” ‘, complete digestion; :. Incomplete digestionri; -_ no digestion.

.icrlr rumttm (rut -u t R/imlororrus mil/impolis. Rlimltxut'rut sp.
y\'le|ds were estimated by spL'Ltropliotontetry.

this method yielded DNA that was less pure. The DNA frag-
ments fro om all soils were larger than 20 kb

A larger- scale gel- plus— —column method was also evaluated
on these soils. The A “MN, and Amy/1.“, ratios of DNA
purified by the large scale method were 1.6 to 18 and >19.
respectively indicating that the DNA was of good quality. For
most of the extracts 01 1.1.1 aliquots resulted in bet tter PCR
amplificatione than did 1- tLl aliquots (Fig 4) eWV soil
extract still appeared to contain substances that interterled with
the reaction because no amplification was obse rvwed il- LLl
aliquots. DNA fragments purified by the large- -scale procedure
were all larger than 20 ltb (data not shown). However, the
large-scale purification method gave poorer DNA recoveries
(27 to 80%; Table 3) than the small- scale methods d.id Recov-
eries were improved slightly (6. g" from 53 to 75" for NK soils
and from ()8 to 85‘" r for CK soils) when DNA Cleanup resin or

Maxipreps Plasmid Purification resin (Promega Corp.w) as
used instead of the Minipreps resin The resin chotce may have

kb 1

234567891011

 

FIG 2 Agarose gel electrophoresrs til loS rDNA amplification products
d NK unl samples Lanes 1. Hmdl ll- ErnR ind

fro mDNA fit me L-Ld
Brmilll- -cut it tctenoplmge lttntliiln molecular m e marker

r11 rig) .. C DNA
tract s .i rid 4 un nltliti d 1 nd 11) ’dtlu tcd (41- -ti ) DNA extracts
putiliL d by LI plus centrifugal Lon \l itralor; 5 and (i. undiluted (RS-rig) :tnd

Eh
ltl 'dilutcd (\ S-n rig) DNA (tracts purified by: . r and 5.
undiluted (311- rig) and 1(1’ -di|nlc dt lln NA extracts purified by single
minicolumn 9 and 10 undiluitrl (25‘ iig) utid lll"- diluted (Zn'vng) DNA ctr
tracts purified cby two \uCL‘L'SStVC minicolumn). ll reaction mixture only wrtliout
DNA mlap

cl plus mll’llClVillmn'

121

VOL.62.1996
Id) 1 2 3 4 5 6 7
d
157“
55 .. fl
as

3 Autoradiogram of hybridization signals with the :5ch gene after
Sou rn transler of rDNA from seeded slertlc N
bacteriopha ge lambda molecular size marker (1 ug): 3. pure-culture DNA 3.
en tcid DNA exit: ICL\. 4, DNA extracts purified by gel electrophoresis plus cen-
trifugal Lonccntraior 5 DNA cx xiracts pit itilied by ge e1 elect trophorcsis pliis col-
umn (i. DNA extracts purified by single column. 7 DNA extracts purified by
double colu urini

to be optimized when recovery of large amounts of DNA is
important.

DISCUSSION

We divided the problem of DNA recovery from soil into two
component methods. i.e.. (i) cell lysis and extraction of crude
DNA and (ii) purification of crude DNA. since there are ad-
vantages in combining different lysis and purification methods
for different cases. The component methods were first devel-
oped on a standard soil (NK) and then tested on a set of more
challenging soils. Three evaluation approaches were used to
determine whether the SDS- based extraction method recov-
ered most of the bacterial DNA. i) DNA recovery efficiency
from seeded bacteria in a standard soil; (ii) lysis efficiency of
indigenous bacteria in more challenging soils; and (iii) com—
parison of crude DNA yields with expected DNA yields.

The SDS-based extraction method resulted in 92 to 99%
recovery of the DNA from bacteria added to soil. These effi-
ciencies were comparable to or higher than those obtained by
other laboratories (6. 21). This first approach to evaluating
DNA recovery, however. can overestimate extraction effi-
ciency. because indigenous bacteria may be more difficult to
lyse than seeded bacteria. When we determined net losses in
indigenous cell counts after extraction, lysis efficiencies varied
from 26 to 92% among five test soils. The variation in cell lysis
apparently reflects differences in soil characteristics and bac-
terial community composition (i.e.. soils exhibiting low cell
lysis may have contained higher proportions of gram-positive
cells). Crude DNA yields from the SDS-based extraction
method agreed reasonably well with expected yields based on
direct microscopic counts. We compared each experimental
yield with a range of expected yields. rather than a single value.
because of the uncertainty regarding the choice of an average
cellular DNA content for soil bacteria

SDS has been the most widely used cell lysis treatment for
DNA extraction from pu ure cultures. soils. an d sediments.
Trevors et a1. (24) found that the SDS- based cell lysis protocol
provided the highest DNA yields in comparison with freeztng-
thawing and Sarkosyl-based lysis protocols. More et a1. (13)

DNA RECOVERY FROM SOILS 321

showed that the percentage of indigenous Cells remaining after
SDS treatment of a sediment (13%) was lower than the per-
centage of cells left after 10 min of bead milling (26%). Our
results indicated that SDS-based cell lysis. in combination with
high-salt treatment and heating. was e ective for most of the
soils but appeared to be influenced by clay content and was not
effective for at least some gram-positive bacteria. Thus. for
soils exhibiting poor cell lysis or studies depending on extensive
sampling of gram-positive DNA. other lysis treatments or comA
binations of treatments could be considered. Combining SDS
with bead mill homogenization resulted in higher cell lysis
efficiency for Baal/us endospores (13). However. bead mill
homogenization and other phystcal methods such as sonication
generally cause severe DNA shearing (ll. 14). Our results
showed that the combination of grinding. freezingthawing.
and SDS resulted in much higher DNA yields from most of the
gram- positive bacteria but without severe shearin

Although DNA purified by all methods could be amplified
and hybridized. some variation in DNA Aup rity was observed
with respect to restriction enzyme digestion (Table 6). The
gel plus- minicolumn method appeared to result in the purest
DNA. because the DNA was completely digested by all en-
zymes examined; the method also provided good recovery with
the standard soil. However. the largebscale gel-plus-column
method gave variable DNA recovery from crude extracts from
other soils. Single— or double-minicolumn purifications ap-

eared to give DNA which was incompletely digested and less
suitable for PCR amplification. as well as having lower recov-
ery efficiency. Single. or double-minicolumn methods, how-
ever. were very rapid and less expensive.

The gelAplus-column method gave very pure DNA. while the
gel-plus-concentrator method gave the highest recovery. The
latter method. however. may not remove all humic contami-
nants from crude extracts of soils with low chromas (0 or 1).
because these soils appear to contain higher proportions of
high-molecular-weight humic acids. 1f gel-plus-concentrator

 

FIG. 4 Agarose gel electrophoresis of 165 rDNA amplification products of
DNA iempl fro ifs purified bty the lirge exc ilc proctdure. Lani: s: i
Hiridlll ELan. and HartiHl-cut bacteriophage1.imbxl.iniolcc.:ilir ~th marker (1

ugh! and 3. undiluted (15‘ ng) an rid ll) 'dilttLdi113T c.1racts. 4 .in l i
undiluted (21) rig) and 10 ‘-diluti:(l (Ill- rig) LP extracts t .in id
(ind It) ' diluted (ll K5 ng) ME ex tirictx; 5 IN” indilutLd (Jri- rig) and
in n" diiuicdtir-i rig) RU- mam in ind 1|.tiltdlluled isnvi mil in ' Illtll a
(lid rig) VH extracts. 12 and 13 tllldllllllcd Idll rig) in nd ll) '-di|ulcd (4- ng)W
extracts; 14 re. tLiion mixture only wrihoui DNA thpI.tLt

122

3

IO
[U

ZHOU ET AL.

purification is used on these soils. concentrator units with the
highest-molecular-weight cutoffs should be used. because some
humic acids have molecular weights of 100,000 or greater.
Larger-scale gel-plus—column purification can provide larger
amounts of DNA but is more expensive. requiring additional
agarose and DNA-binding resin.

In summary, the DNA extraction and purification methods
evaluated here are simple, rapid, and efficient for most soils
and purposes. DNA could be extracted from eight soil samples
in 6 h by the SDS-based method. Because of the gentle nature
of the extraction treatment, the DNA fragment size in crude
extracts was >23 kb. DNA purification required 2 to 4 h for the
single- or double-column method. 8 to 10 h for the gel-plus-
column method. and 12 to 14 h for the geLplus-concentrator
method. If DNA purity is of the greatest concern, we recom-
mend gel—plus-column methods. It is also important to recog-
nize that no single method of cell lysis or purification will be
appropriate for all soils and experimental goals. The basic
methods suggested should be appropriate for the more com-
mon cases. but different combinations and modifications of
lysis and purification protocols will probably be needed for
some conditions.

ACKNOWLEDGMENTS

This project was supported by National Science Foundation (NSF)
grant BIR9120000 to the Center for Microbial Ecology and by Na-
tional Institute of Environmental Health Sciences Superfund Research
and Education grant ES-049ll. A portion of the research was also
supported by an LTER graduate research grant. NSF grant
DEB9211771.

We thank David Harris for valuable suggestions. technical advice.
and assistance with microscopy and image-processing analysis: Delbert
Mokma for soil particle size determmations: and Greg Thorn for
helpful suggestions.

REFERENCES

1. Ausubel. F. M. R. Brent. R. E. Kingston. D. D. Moore. J. G. Seidman. J. A.
Smith. and K. Struhl (ed.). 1990. Current protocols in molecular biology. vol.
2. Greene Publishing Associates and Wiley-lnterscience. New York.

. Baldten. L R.. and R. A. Olsen. 1989. DNA content of soil bacteria of
different cell size. Soil Biol. Biochem. 21:789-793.

3. Bloem. J.. P. R. Bolllnis. M. R. Veringa. and J. Weiringa. 1995. Microscopic

methods for counting bacteria and fungi in soil. p. 102-173. In K. Alet‘ and
P. Nannipieri (ed). Methods in applied sod microbiology and biochemistry.
Academic Press Ltd. London.

4. Day. P. R. 1965. Particle fractionation and particle size analysis. p. 545-567.
In C. A. Black (ed). Methods of soil analysis. vol. 1. American Society of
Agronomy. Madison. Wis.

5. Dom. E. M. Hellwig. W. Reinelte. and H. Knackmnss. 1974. Isolation and
characterization of a 3-chlorobenzoate degrading pseudomonad. Arch. Mi-
crobiol. 99:61-70.

6. Erb. R. W.. and l. Wagner-Dobler. 1993. Detection of polychlorinated bi-
phenyl degradation genes in polluted sediments by direct DNA extraction
and polymerase chain reaction. Appl. Environ. Microbiol. 59:4065—4073.

. Food and Agriculture Organiaation of the United Nations—United Nations

OJ

\l

‘O

16.

I7.

APPt. ENVIRON. Mtcaoarou

Educational. Scientific and Cultural Organization. 1988. Soil map of the
world—revised legend World Soil Resources report 60. Food and Agricul-
ture Organization. Rome.

. Fry. J. C. 1990. Direct methods and biomass estimation. Methods Microbiol.

22:41-85.

. Fulthorpe. R. R... A. N. Rhodes. and J. M. Tiedje. Pristine soils mineralize

3-chlorobenzoate and 2.4-dichlorophenoxyacetate via difl’erenl microbial
populations. Submitted for publication.

. Holben. W. E. J. K. Jansson. B. K. Chelm. and .I. M. Tiedje. 1988. DNA

probe method for the detection of specific microorganisms in the soil bac-
terial community. Appl. Enx'iron, Microbiol. 54:703—711.

. Uesacli. W.. and E. Stackebrandt. I992 Occurrence of novel groups of the

domain Bacteria as revealed by analysis of genetic material isolated from an
Australian terrestrial environment. J. Bacteriol. ”45072-5078.

. Liesack. W.. H. Weyland. and E. Stackebrandt. 1901. Potential risks of gene

amplification by PCR as determined by 165 rDNA analysis of a mixed-
culture of strict barophilic bacteria. Microb. Ecol. 21:191-198.

. More. M. L. J. B. Herrick. M. C. Silva. W. C. Ghiorse. and E. L Madsen.

1994. Quantitative cell lysis of indigenous microorganisms and rapid extrac-
tion of microbial DNA from sediment. Appl. Environ. Microbiol. 6021572-
1580.

. Ogram. A.. G. S. Sayler. and T. Bark-y. 1987. The extraction and purification

of microbial DNA from sediments. J. Microbiol. Methods 7:57—66.

. Porteons. L A., and J. L Amstmng. 1991. Recovery of bulk DNA from soil

by a rapid. small-scale extraction method. Curr. Microbiol. 22:345—348.
Sher-r. B. F. E. B. Sherr. and R. D. Fallon. 1987. Use of monodispersed.
fiuorescently labeled bacteria to estimate in situ protozoan bactivory. Appl.
Environ. Microbiol. 53:955—965.

Simooet. P.. M.-C. Grosjean. A. K. Misra. S. Nazaret. B. Connoyer. and P.
Normand. 1901. Frankie genus-specific characterization by polymerase chain
reaction. Appl. Em'iron. Microbiol. 573278—3286.

. Smalls. K- N. Cresswell. L C. Mendonca-Hagler. A. Wolters. and J. D. van

Elsas. 1903. Rapid DNA extraction protocol from soil for polymerase chain
reaction-mediated amplification. J. Appl. Bacterial. 74:78—85.

. Stefan. R. J.. and R. M. Atlas. 1988. DNA amplification to enhance detec-

tion of genetically engineered bacteria in em'ironmental samples. Appl.
Environ. Microbiol. 542155-2191.

Stefan. R. J., J. Golcsoyr. A. K. Bej. and R. M. Atlas. 1988. Recovery of DNA
from soils and sediments. Appl. Environ. Microbiol. 54:2“)8—2915.

. Tebbe. C. C.. and W. Vahjen. 1993. Interference of humic acids and DNA

extracted directly from soil in detection and transformation of recombinant
DNA from bacteria and yeast. Appl. Environ. Microbiol. 593657—2665.

. Torsvilt. V.. J. Goltsoyr. and F. L. One. 1900. High diversity in DNA of soil

bacteria. Appl. Environ. Microbiol. 56:782—787.

. Torsvilt. V. L. and J. GM. 1978. Determination of bacterial DNA in soil.

SOiI Biol. Biochem. 10:7—12.

. Trevors. J. ’l‘.. H. Lee. and S. Coolt. 1992. Direct extraction of DNA from soil.

Microb. Releases l:IlI-ll5.

. Tsai. Y.-L. and B. H. Olson. 1992. Rapid method for separation of bacterial

DNA from humic substances in sediments for polymerase chain reaction.
Appl. EnViron. Microbiol. 582292-2295.

. Viles. C. L. and M. E. Sieraclti. 1992. Measurement of marine picoplankton

cell size by using a cooled. charge—coupled device camera Wllh image-antic
Iyzed fluorescence microscopy. Appl. Environ. Microbiol. 58:584—592.

. Ward. D. M.. R. Weller. and M. M. Bateson. 1990. 168 rRNA sequences

reveal numerous uncultured inhabitants in a natural community. Nature
(London) 345:63—65.

Zhou. .I.-Z.. M. R. Fries. J. c. Cine-shim. and J. M. Tiedje. I995. Phy~
logenetic analyses of a new group of denitritiers capable of anaerobic growth
on toluene: description of Azoarcw tolulyticus sp. nov. Int. J. Syst. Bacteriol.
45:500—506.

. Zhou. J.-Z.. and J. M. TIedje. 1995. Gene transfer from a bacterium injected

into an aquifer to an indigenous bacterium. Mol. Ecol. 4:613—618.

"Illlllllllllllllllllll“