Shaylie Augustine
California University of Pennsylvania
Honors Advisory Board Committee:
Dr. Craig Fox
Dr. Michelle Valkanas
Dr. Min Li
Dr. Peter Cormas
Keywords: Donora Smog, Donora Zinc Works, zinc production, zinc resistance, soil bacteria, zinc,
heavy metal pollution, organic matter, moisture, growth curves, genome assembly, Serratia
liquefacines, Bacillus cereus, Alcaligenes faecalis, Delftia acidovorans. Stenotrophomonas
maltophilia

ACKNOWLEDGEMENTS

The following individuals are acknowledged by the author for their significant
contributions to this study: Dr. Robert Whyte who provided assistance and supervision in
processing soil samples for moisture and organic content, Dr. Min Li who provided assistance
and supervision in processing soil samples for zinc content, and Dr. Peter Cormas who acted as
an Honors Advisory Board committee member.
The author would also like to acknowledge administration of the Honors Program not
only for overseeing this project, but the entirety of the undergraduate’s studies: Dr. Craig Fox,
Associative Director, and Dr. Mark Aune, Director.
Gratitude is extended to James Augustine, Sonya Augustine, Carlene Pugh, Austin Smith,
and Madeline Sanders for their constant motivation and encouragement over the course of the
study.
A special acknowledgement is extended to Dr. Michelle Valkanas, the advisor of this
study. Her first-hand experience and knowledge of soil microorganisms, as well as the
instruction and supervision she provided over a multitude of experiments and the many written
drafts, combined with her endless support and encouragement contributed significantly to the
overall project. Without her time, motivation, and instruction, this study would most certainly
not have produced such a wonderful product. Her selflessness and dedication are greatly
appreciated and will never be forgotten.

TABLE OF CONTENTS
Acknowledgements…………………………………………………………………….………………………………….…………..ii
List of Tables………………………………………………………………………..………………………………………………..…..iv
List of Figures…………………………………………………………………………..…………………………………………………v
Abstract………………………………………………………………………………..……………………………………………………vi
Introduction……………………………………………………………………………..………………………………………………..1
Materials and Methods………………………………………………………………………………………..…………………….4
Results…………………………………………………………………………………………..…………………………………………...7
Discussion………………………………………………………………….……………………………………………………………..13
Conclusion…………………………………………………………………….………………………………………………………….19
References………………………………………………………………………………………………………………………………..19
Appendices……………………………………………………………………………………………………………………………….22

iii

LIST OF TABLES
Table 1: Zinc Content in the Soil………………………………………………………………………………..……………7
Table 2: Results from DNA Extraction………………………………..………………………………………………….10

LIST OF FIGURES
Figure 1: A map of Donora, PA……………………………………………………………………………………………..…1
Figure 2: American Steel and Wire - Donora Wire Works and Zinc Works…………………………….…2
Figure 3: A mill worker stands behind newly minted zinc ingots……………………………………………..2
Figure 4: The calculated average moisture content of three samples from 7 experimental
sites………………………………………………………………………………………………………………………………………..7
Figure 5: The calculated average organic content of three samples from 7 experimental
sites……….…………………………………………………………………………………………………………………………….…7
Figure 6: Control 2 Isolate Zinc Growth Curves…………………………………………….………………………….8
Figure 7: Tryptic Soy Media Growth Curves for Donora, Pollution, and Inorganic Farm
Sites……………………………………………………………………………………………………………………………………….9

ABSTRACT

THE DEEP-ROOTED DAMAGE OF THE DONORA SMOG DISASTER

By
Shaylie Augustine
April 2022

Honors thesis supervised by Dr. Michelle M. Valkanas
The Donora smog is an infamous event caused by Donora Zinc Works, which operated
for thirty-three long years prior to the disaster which resulted in a foggy haze mixed with
atmospheric pollutants to smother the proximal areas. The Donora smog occurred over the
Donora and Webster areas on October 29th, 1948. Soil samples were collected, in triplicate, from
the location where the plant used to stand, as well as from surrounding areas not impacted by
industrialization. The soil samples were analyzed for pH, moisture content, organic matter
content, and zinc concentrations. Bacteria isolated from the soil were tested for their ability to
grow in high zinc concentrations and DNA was extracted from isolates exhibiting zinc resistance.
The extracted DNA underwent shotgun sequencing to generate genome assemblies of the
isolates. Some of the genomes identified were Serratia liquefacines, Alcaligenes faecalis, and
Stenotrophomonas maltophilia which have previously shown zinc resistance. The soil samples
collected from Donora displayed high concentrations of zinc and the microbes isolated from the
samples displayed the ability to grow in high zinc concentrations. The results from this study
provide evidence to support that there are long-lasting effects of industrialization on the
bacterial communities within the soil.

1. Introduction

Known for its protective properties, a
component of brass zinc was used in
manufacturing cartridges, shells, fuses, and
detonators. When alloyed with aluminum,
magnesium, and manganese, it was used in
the manufacturing of shafts, propellers, and
bearings of aircraft parts as well as marine
hardware, cables, canisters, and drums for
the Navy (Donora Historical Society and
Smog Museum).

1.1 Donora Smog Background
A small town located south of Pittsburgh,
settled along a horseshoe bend in the
Monongahela River is known as Donora, and
the town across the river, Webster (Figure
1). Two major plants can be seen pictured in
Figure 2, the American Steel and Wire plant
(to the left), and the Donora Zinc Works (to
the right), which were found nestled in the
small towns and the air pollution they
produced was nothing new to its residents.
The infamous Donora Zinc Works would
begin construction in 1915, and on October
29th, and the plant would produce its first
Zinc, running until its closure in 1957
(Donora Historical Society and Smog
Museum).

Shortly after production began, the
residents of Donora and Webster started to
notice a difference in the atmosphere,
especially the land-owning farmers. At the
beginning of the 1920’s residential farm
owners sued for damages such as loss of
crop, livestock, topsoil, and even destruction
of fences and houses. During the height of
the Great Depression, many local residents
came together to sue Donora Zinc Works for
damages to their health due to the
atmospheric effluent (Snyder, 1994).

One of the largest of its time, the Donora
Zinc Works sat on 4,000 feet of land and
played a vital role in the National Defense
efforts of World War II. Sulfuric acid, one of
the by-products from the zinc plant was used
in manufacturing explosives, however the
cadmium and zinc by-products were less
useful.

For decades, Zinc Works was viewed as a
threat to public health, and that would
become undeniably true in October 1948.
On Tuesday, October 26th, 1948, the fog from
the industrial production lingered longer
than usual. By Wednesday, it had become
more noticeable, and did not burn off at all
by Thursday (Snyder, 1994). When Friday
evening arrived, visibility was dwindling, and
many elderly residents reported signs of
respiratory distress. Volunteer firefighters
navigated the reluctant fog on foot to
administer oxygen to elderly residents and
asthmatics. Despite this, Donora citizens did
not cancel neither their annual Halloween
parade on Friday, October 29th, 1948, nor a
high school football game the following day.
As the smog got thicker, more citizens began
to exhibit signs of respiratory distress, and as
if from the plot of a horror movie, on
October 30th, 1948, at around 2:00 AM, the

Figure 1: A map of Donora, PA Source: (Ivel).

Figure 2: American Steel and Wire - Donora Wire Works and Zinc Works. The photo above shows the American
Steel and Wire (left) and Donora Wire Works and Zinc Works (right) in 1930s as taken from across the Monongahela
River in Webster. Source: (Donora Historical Society and Smog Museum).

smog took its first victim. Twelve hours later,
another 17 Donora and Webster residents
would be pronounced dead. Of the 14,000
Donora residents, the number of deceased
ranges from 20-26 (Jacobs et al., 2018;
Donora Historical Society and Smog
Museum). Many other individuals were
impacted with 1,440 individuals reporting to
have suffered from serious illness and 4,470
from moderate symptoms (Jacobs et al.,
2018). These numbers of course do not
include individuals who suffered from long
term effects, or died within the following
weeks, but display the volume of destruction
caused by the smog. The mills did not cease
production until Sunday, October 31st, 1948,
at which point a heavy rain had begun to
dissipate the lethal smog. (Baranauskas,
2017).

nitrogen dioxide, and multiple sulfur
containing compounds as well as heavy
metal particulates (Jacobs et al., 2018).
The preliminary report listed the other
contributing factors to be the unusual
weather system and the geography of
Donora that aided in the smog formation.
The temperature inversion caused the warm
air and smog to be trapped below the cold
air between the high mountains of the
horseshoe bend in the Monongahela River
(Jacobs et al., 2018).
This investigation and its findings
contributed to the implementation of the
1955 Air Pollution Control Act, and
eventually the 1970 Clean Air Act which
would authorize the development of federal
and state regulations that limit the emissions
of industrial and mobile sources of air
pollution (Helfand et al., 2001).

After the poisonous gases cleared and the
town quieted back down, an investigation
was conducted by the United States Public
Health Service to determine the cause of this
smog (Jacobs et al., 2018). A preliminary
report of the investigation released in 1949
concluded that the smog was caused by a
combination of three factors. The first, and
probably most obvious contributor, was the
air pollution emitted by both the American
Steel and Wire company and the Donora Zinc
Works, with Zinc Works being named as the
major polluter due to the emission of
hydrogen fluoride, carbon monoxide,

Figure 3: A mill worker stands behind newly minted
zinc ingots Source: (Bleiwas & DiFrancesco, 2010).

After the report was released to the public
by the United States Health Public Health
Service, many individuals began to identify
gaps within the investigation and other longterm health effects that had not been
accounted for. An editorial published by the
New England Journal of Medicine described
the United States Public Health Service
report as a missed opportunity for
conducting more detailed research of health
effects on long-term pollution and severe
acute effects (Jacobs et al., 2018).

Zinc is an essential element biologically
aiding in many crucial processes such as DNA
and RNA metabolism, protein processing,
and neural response modulation. Like most
biologically essential elements, there are
serious detrimental effects at decreased or
heightened levels of zinc ranging from
diarrhea and urinary problems to
neurological problems, organ damage and
genetic defects (Environmental Pollution
Centers, 2022).
Unnaturally occurring zinc like that
accumulated from zinc mining and
industrialization can easily contaminate soil
and water. There are two processes
available for zinc manufacturing and both
begin with roasting or sintering to remove
the sulfur from the mined zinc (zinc blende),
producing zinc oxide and releasing sulfur
dioxide as a by-product (Greenspec, 2022).

This event is known today as the Donora
smog and since then much research has
been conducted on the surrounding areas to
further discover the negative effects that the
smog has caused. In 1961, two
biostatisticians from the University of
Pittsburgh conducted a study and found that
there was a higher-than-expected mortality
rate for cardiovascular disease and cancer in
Donora during the decade following the
smog (Jacobs et al., 2018).

During the hydrometallurgical process the
zinc oxide is then separated from other
calcines using sulfuric acid which dissolves
the zinc leaving iron, lead, and silver
precipitates that are later removed using
zinc dust. During the pyrometallurgical
process the zinc oxide is mixed with crushed
coke and heated at extreme temperatures to
reduce the zinc oxide to metallic zinc. The
metal is condensed leaving behind cadmium,
lead, and iron impurities (Greenspec, 2022).

An additional study published in 2017 details
the results found from sediment cores
retrieved from a lake 6-miles northeast of
Donora. The cores showed that after the
opening of Zinc Works in 1915 there was a
substantial increase in cadmium, lead, and
zinc levels, all known carcinogens, and the
levels did not subside following the closing of
the plant in 1957. The cadmium and lead
contaminants were found to have remained
at
a
higher
than
recommended
concentration for 70 years following the
smog event. It was noted that the
disturbance of the soil can release these
contaminants back into the water, and
because of this, the pollution from the event
remains a risk, even today (Rossi et al.,
2017).

Zinc industrialization has many negative
effects on the environment including the
production of sulfur dioxide which can cause
acid rain. Other unwanted byproducts that
are produced during the manufacturing
include cadmium vapor, sulfur oxide, carbon
monoxide, carbon dioxide, and other heavy
metals. Many of the vapors released have
negative effects like carbon monoxide which
is known to be ozone forming. Many sites of
previous zinc mining or manufacturing have
been found to leach significant amounts of

1.2 Effects of Zinc Industrialization on the
Environment
3

zinc, cadmium, and other heavy metals into
the soil and surrounding waterways, posing
a health threat to humans and the
environment (Greenspec, 2022).

isolated from the soil. If the industrial
pollution in this area still exists today, then
the soil collected from Donora will contain a
higher zinc concentration than that collected
from the other test sites. It is also
hypothesized that bacteria isolated from the
Donora site will exhibit a greater ability to
grow in high zinc concentrations when
compared to other test sites.

1.3 Microorganisms and Zinc Pollution
Zinc plays a vital role in plant nourishment at
low levels. At higher concentrations zinc can
become phytotoxic and persist for a long
period of time, unlike other pollutants that
can be chemically and biologically degraded
(Kour et al., 2020). Bioremediation is a
process that uses living organisms, such as
soil bacteria, to detoxify heavy metals from
the environment (Kour et al., 2020).

2. Materials and Methods
2.1 Soil Collection
Soil samples were collected in triplicate at
seven
sites
within
southwestern
Pennsylvania to explore the metabolic
capabilities of the microorganisms within
them. The sites were chosen based on their
proximity and relationship to the location
where the Zinc Works production plant was
known to reside before its demolition.

High concentrations of zinc within the soil
effects the flora and fauna that inhabits it.
Excess zinc can alter plant development,
although a few species of plants have
developed the ability to grow in high zinc
concentrations. The effect that high levels of
zinc has on bioavailability of plants and soil
bacteria depends on a combination of
factors including, but not limited to,
microbial community structure and available
organic matter (Balafrej et al., 2020).

Soil collected from Donora Industrial Park
(40°10’41” N 79°51’11” W) is referred to as
the “Donora” site and represents soil directly
affected by the Zinc Works production. Soil
from Nemacolin Park (39°52’44” N
79°55’13” W) in Carmichaels, PA which is
located downhill from Hilltop Energy Center
and was collected to represent soil affected
from a currently active powerplant and is
referred to as the “Pollution” site. Soil
collected from California University of
Pennsylvania’s SAI Farm (40°02’50” N
79°54’49” W) from both the organic orchard
and organic produce plots were chosen to
study the agriculture located near the
affected area and are referred to as “Organic
Orchard” and “Organic Farm”, respectively.
Additional samples from inorganic orchard
and produce plots (40°04’48” N 79°08’20”
W) in Somerset, PA were obtained to study
an agriculture environment distant from the
affected area and are referred to as
“Inorganic Orchard” and “Inorganic Farm”,

High concentrations of zinc in soil kills the
majority of the microflora, creating a
selective pressure for the emergence of
heavy
metal
resistant
strains
of
microorganisms. Heavy metal resistant
microorganisms effect the mobility the
heavy metals and detoxify the soil by
converting the toxic form of heavy metals
into a nontoxic form through the production
of metabolites (Kour et al., 2019).
Donora is home to a monumental event
caused by zinc production, and because of
this, was chosen as the target area of study.
Soil was collected from Donora and
surrounding areas to analyze the zinc,
moisture, and organic matter content, as
well as the capabilities of microorganisms
4

respectively. Finally, samples were obtained
from an old forest located on a hill above the
SAI Farm (40°02’40” N 79°54’51” W)
representing the control group and is
referred to as “control” (Supp. Table 1).

was loaded, the samples were heated for
approximately 24 hours at 350°C, then
removed, reweighed, and recorded. The
organic matter content was calculated by
first finding the percent of mineral content
then subtracting from 100.

On May 17th, 2021, samples were collected
from the Donora, Pollution, SAI Orchard, and
SAI Farm sites, on May 20th, 2021, samples
were collected from the Organic Orchard
and Organic Farm sites, and on May 24th,
2021, samples were collected from the
Control site (Supp. Table 1).

2.3 Bacterial Isolation and Broth Cultures
Soil bacteria was cultivated using three types
of agars to diversify the organisms collected.
Lima bean, tryptic soy, and nutrient agar
(Fisher Bioreagents) plates were made, and
the soil samples were swabbed with a sterile
cotton swab and plated onto the prepared
plates. Individual colonies of unique colony
morphology were selected for additional
screening and were restreaked in order to
purify. Once the bacteria were purified, they
were transferred from the plates to a liquid
broth culture of the appropriate medium
using aseptic technique.

The soil was collected using a pre-sterilized
soil core and 5% Lysol was used to clean the
device between sites to ensure that there
was no cross-contamination between the
samples obtained from each site. The soil
samples were refrigerated in storage
between tests.
2.2 Moisture and Organic Content Test

2.4 Zinc Content Test

Soil samples were weighed and dried to
calculate moisture and organic content. The
moisture content test was conducted by
weighing at least 5 grams but not exceeding
28 grams of soil from each sample and
placing it in an aluminum foil boat, both the
weight of the foil and the wet weight of the
soil were recorded. The samples were then
placed into a dehydration machine located
in the biology department of California
University
of
Pennsylvania.
After
approximately 24 hours at 105°C the
samples were removed from the dehydrator
and weighed again, then the moisture
content for each sample was calculated by
first finding the percent soil content then
subtracting from 100.

Soil samples from the Donora, Pollution, and
Control sites were processed for zinc
concentration by first drying the soil samples
for 48 hours at 65°C. The three samples that
were collected for each site were combined
to provide better quality data, and
approximately five grams of dried soil from
each site were digested using 20 mL of nitric
acid, 50 mL of hydrochloric acid, and 4 mL of
hydrogen peroxide. The solutions were then
filtered through a 12.5 cm diameter filter
paper using gravity filtration and the filtrate
was added to deionized water to achieve a
diluted solution of 250 mL. Zinc calibration
standards of 0.1 PPM, 0.5 PPM, and 1 PPM
were prepared from a 1000 PPM zinc stock
solution for a total volume of 50 mL. The
prepared solutions were then analyzed using
an Atomic Absorption Spectroscopy (AAS)
located in the Chemistry Department at
California University of Pennsylvania. The

Similarly, the organic content was tested
using the dehydrated soil from the moisture
test. Approximately 5 grams of soil was
weighed, added to a pre-weighed crucible,
and arranged in the furnace. After the oven
5

zinc content of the samples was determined
using the average of three reads from each
sample and the results from the calculation
were graphed.

The extracted DNA samples were sent to the
Microbial Genome Sequencing Center
(Pittsburgh, PA) where shotgun sequencing
was performed on the samples. The FASTQ
files received back from sequencing were
scaffolded, assembled, and annotated for
each genome using Kbase data base
(https://www.kbase.us/). The sequences
were imported into FASTQ files to get a
paired-end library of each sample. The
FASTQC report written by Simon Andrews of
Babraham Bioinformatics was then
conducted on the samples to assess the read
quality. From this quality control check the
per base sequence quality, or Phred score,
nucleotide distribution, unidentified bases
and adapters, and Kmer content were
analyzed to assess the quality of the pairedend library. The trimmomatic program,
written by Anthony Bolger, Marc Lohse, and
Bjoern Usadel was used on each sample to
remove any poor-quality sequences, using
the results from the Kmer data to determine
the Head Crop length adjustments. The
trimmomatic program parameters used
were a sliding window size of 4 and a sliding
window minimum quality of 15. An
additional FASTQC report was obtained, and
the results were compared to those of the
first report. The reads were then assembled
through de novo assembly using both the
Velvet and SPAdes assembler. A QUAST
report was then conducted on both the
Velvet and SPAdes assemblies and the
reports were compared to determine which
assembly would be used moving forward.
The chosen assemblies were then annotated
using the annotate assembler Annotate
Microbial Contigs using Rapid Annotations
using Subsystems Technology (RAST). The
quality of the assembly was then analyzed
using the CheckM tool. The RAST annotated
genome assembly for each sample was then
used to build a metabolic model using the

2.5 Zinc Growth Curves
Isolates from the Donora, Pollution, Control,
and an isolate from the Inorganic Farm
(n=28) were tested on their ability to grow in
high zinc environments. The Inorganic Farm
isolate was included due to the isolates
ability to emit a purple color that was not
observed by any of the other bacterial
communities. A 96 well plate was used to
simulate microenvironments which were
then observed using a plate reader (BioTek
Instruments Inc). The absorbency was
measured at 600nm every 3 hours for 27
hours. The isolates were exposed to 80 mM,
40 mM, 20 mM, 10 mM, 5 mM and 0 mM
concentrations of zinc with their
corresponding medias using 20 μL of growth
obtained from the broth cultures for a final
volume of 220 μL. Wells containing only
media and media plus zinc only were created
as controls to ensure there was no
contamination. Each sample was tested in
duplicate, including controls, to ensure the
quality of the data collected. Raw data for
the zinc growth curves can be found in Supp.
Tables 7-14.
2.6 DNA Extraction
Eight isolates: C2-NB, C2-LB, P1-TSB, C2-TSB,
D1-TSB, D2-TSB, D3-TSB, and IO-TSB that
displayed zinc resistance were selected for
DNA extraction. DNA extraction was
completed using a DNeasy® UltraClean®
Microbial Kit (Qiagen, Hilden, Germany). The
extracted DNA was analyzed using a Broad
Range Assay kit on a Qubit 4 Fluorometer
(Invitrogen, United States) to determine the
DNA concentration of each sample.
2.7 Genome Assembly
6

Build Metabolic Model tool. The Insert
Genome into Species Tree tool was used to
determine the identity of the assembled
genomes
and
compare
taxonomic
relationships. The 16S gene nucleotide
sequence was then compared using another
nucleotide sequencing database, BLAST
(https://blast.ncbi.nlm.nih.gov/Blast.cgi) to
confirm the identities of the isolates

Moisture Content (%)

25
20
15
10
5
0
Control

Donora

Pollution

SAI
Orchard

SAI Farm Inorganic Inorganic
Orchard
Farm

Sites

3. Results

Figure 4: The calculated average moisture content of
three samples from 7 experimental sites. The red line
indicates optimal agricultural moisture content
(Blumberg, 1982).

3.1 Moisture and Organic Content Test

Organic Matter Content (%)

The results from the moisture content test
are the average of the three samples from
each site (Figure 4 and Supp. Table 1). The
results revealed that all the samples had
similar calculated moisture percentages. The
Pollution site was determined to have the
lowest content of moisture with 19.48% as
the calculated average percent, and the
Donora site had the highest, with a
calculated average of 23.56%.
The results from the organic matter content
are displayed as the average of the three
samples from each site (Figure 5 and Supp.
Table 2). The results reveled that overall, the
samples collected from the Organic SAI Farm
had the lowest amount of organic matter,
the SAI Farm site was determined to be
4.84% and the SAI Orchard site, 5.56%. The
Control site was determined to have the
most organic matter with a recorded
average of 19.71% organic matter.

25
20
15
10
5
0
Control

Donora

Pollution

SAI
Orchard

SAI Farm Inorganic Inorganic
Orchard
Farm

Sites

Figure 5: The calculated average organic content of three
samples from 7 experimental sites. The red line indicates
optimal agricultural moisture content (Blumberg, 1982).

Site
Name
Control

Zinc Content (PPM)

The zinc content for the control site was the
lowest of the three sites tested and had a
zinc concentration of 3.945 PPM (Table 1).
The soil from the pollution site was
determined to contain 21.645 PPM of zinc
(Table 1). The Donora site contained a
substantially higher amount of zinc and had
a zinc concentration of 51.48 PPM (Table 1).

3.945

3.3 Zinc Growth Curves

Donora

51.48

Pollution

21.645

The results obtained from the growth curve
plates were based off the ninth read from

3.2 Zinc Content Test
Table 1: Zinc Content in the Soil

every test (27 hours in total). The
absorbance for each well was averaged with
its duplicate and the results are displayed in
Figure 6 and 7.

2.5

ABSORBANCE

The Lima Bean media showed that at 80 mM
of zinc isolate Control 2 had displayed the
least inhibition when compared to its control
(Figure 6A). It is observed that the isolate
performed better at 80 mM zinc
concentration, having a faster growth rate
than its control from reads 1-6 (Figure 6A).
Because of this, the isolate was selected to
investigate further.

1.5
1
0.5
0
Read Read Read Read Read Read Read Read Read
1
2
3
4
5
6
7
8
9

The nutrient broth media plate for Control 2
shows that the isolate performed better
than its control at 5 mM concentration of
zinc from reads 1-3 (Figure 6B). The results
also show that the isolate had a faster
growth rate at 80 mM than at 40 mM when
compared to its control, and because of this,
was also chosen to investigate further.

C2 80 mM

C2 40 mM

C2 20 mM

C2 10 mM

C2 5 mM

C2 0 mM

2.5

ABSORBANCE

2
1.5
1
0.5

Isolates growing in tryptic soy media grew at
a faster rate in comparison to the other two
medias overall (Figure 6 and 7). Isolate
Control 2 in tryptic soy media displayed
minimal inhibition at 5 mM, 10 mM, and 20
mM when compared to its control. It was
observed that the isolate experienced
inhibition at 40 mM, displaying slower
growth rates for 40 mM and 80 mM
concentrations when compared to its
control (Figure 6C). Due to the minimal
inhibition observed at lower concentrations
the isolate was selected to be used in future
experiments.

0
Read Read Read Read Read Read Read Read Read
1
2
3
4
5
6
7
8
9

C2 80 mM

C2 40 mM

C2 20 mM

C2 10 mM

C2 5 mM

C2 0 mM

2.5

ABSORBANCE

Figure 6: Control 2 Isolate Zinc Growth Curves.
Growth curves for Control 2 in lima bean media
(A), nutrient broth (B), and tryptic soy media (C)
were averaged with their duplicate and the
recorded absorbency was plotted against read
time for a total of 27 hours. The control is
denoted as “C2 0 mM” and can be visualized as
the green curve in figures A-C.

1.5
1
0.5
0

Read Read Read Read Read Read Read Read Read
1
2
3
4
5
6
7
8
9

C2 80 mM

C2 40 mM

C2 20 mM

C2 10 mM

C2 5 mM

C2 0 mM

A 2.5

2.5
2

ABSORBANCE

2
1.5
1
0.5

1.5
1
0.5

0
Read Read Read Read Read Read Read Read Read
1
2
3
4
5
6
7
8
9

Read Read Read Read Read Read Read Read Read
1
2
3
4
5
6
7
8
9

D1 80 mM

D1 40 mM

D1 20 mM

D2 80 mM

D2 40 mM

D2 20 mM

D1 10 mM

D1 5 mM

D1 0 mM

D2 10 mM

D2 5 mM

D2 0 mM

D 2.5

ABSORBANCE

C 2.5

ABSORBANCE

1.5
1

0.5

Read Read Read Read Read Read Read Read Read
1
2
3
4
5
6
7
8
9

1.5

D3 80 mM

D3 40 mM

D3 20 mM

P1 80 mM

P1 40 mM

P1 20 mM

D3 10 mM

D3 5 mM

D3 0 mM

P1 10 mM

P1 5 mM

P1 0 mM

2.5

ABSORBANCE

1.5
1

Figure 7: Tryptic Soy Media Growth Curves for Donora,
Pollution, and Inorganic Farm Sites. Growth curves for
Donora 1 (A), Donora 2 (B), Donora 3 (C), Pollution 1
(D), and Inorganic Farm (E) were averaged with their
duplicate and the recorded absorbency was plotted
against read time for a total of 27 hours. The control is
denoted at “*Isolate abbreviation* 0 mM” and can be
visualized as the green curve in figures A-E.

0.5
0
Read Read Read Read Read Read Read Read Read
1
2
3
4
5
6
7
8
9
IO 80 mM

IO 40 mM

IO 20 mM

IO 10 mM

IO 5 mM

IO 0 mM

Isolate Donora 1 in tryptic soy media
observed inhibition at 40 mM. For
concentrations of 5 mM, 10 mM and 20 mM,
the isolate grew with minimal inhibition,
however, much slower growth rates were
observed for 40 mM and 80 mM
concentration when compared to its control
(Figure 7A). Due to the minimal inhibition at
lower concentrations, the isolate was
selected to investigate further.

Table 2: Results from DNA Extraction

For both isolate Donora 2 and Donora 3
growing in tryptic soy media, a faster growth
rate was observed at 40 mM from reads 1-2
when compared to their controls (Figure 7B
& 7C). Isolate Donora 2 displayed inhibition
at 40 mM (Figure 7B) and isolate Donora 3
displayed inhibition at 20 mM (Figure 7C)
when compared to their controls after 27
hours. Minimal inhibition was observed at
lower concentrations of zinc (5 mM and 10
mM) when compared to their controls and
therefore both isolates were selected for
further testing (Figure 7B & 7C).

Sample Name

Qubit Reading (ng/µL)

TSB C2

15.2

TSB P1

29.6

NB C2

33.3

TSB D1

280

TSB D2

240

TSB D3

33.4

LB C2

TSB Purple

328

3.4 Genome Assembly
All isolate identities were determined by
cross analyzing 16s sequences with the
BLAST database and only those containing a
100% query cover that scored higher than
90% for the percent identity of the genome
were recorded.
The Inorganic Farm genome in tryptic soy
media had 1,984,094 paired end sequences.
Most sequences had a length of 147.68 base
pairs and a mean Phred score of 33.36. After
quality filtering, 644 base pairs were
dropped. The SPAdes assembly contained
longer and higher quality contigs when
compared to the Velvet assembly and was
used for downstream annotation and
analysis (Supp. Figure 1).

Isolate Pollution 1 growing in tryptic soy
media displayed some degree of inhibition at
10 mM, 20 mM, 40 mM and 80 mM
concentrations when compared to its
control (Figure 7D). At reads 8-9 the isolate
performed better than its control at 5 mM
and was selected for further testing.
Isolate Inorganic Farm growing in tryptic soy
experienced inhibition at 40 mM when
compared to its control. At 40 mM the
isolate performed better than its control for
reads 1-2 (Figure 7E). Due to the results
provided from the growth curve, the isolate
was chosen to continue investigating.

The longest contig was 538,056 base pairs
and the total genome size was 51,013,143
base pairs in length with a 58.16% GC
content. The genome was assembled from
2,271 contigs and the genome was
determined to have 100% completeness
based on marker lineage across 5,656
genomes (56 markers).

DNA Extraction
The Qubit results of the extracted DNA can
be found in Table 2.

The Donora 1 genome in tryptic soy media
had 1,759,608 paired end sequences with a
mean read length of 146.44 base pairs. The

genome had a mean Phred score of 33.38,
after quality filtering 591 base pairs were
dropped. The SPAdes assembly contained
longer, higher quality contigs when
compared to the Velvet assembly and was
used for downstream annotation and
analysis (Supp. Figure 2).

to also be Serratia liquefaciens or a close
relative with a 99.92% identity and 100%
query cover.
The Donora 3 isolate cultivated in tryptic soy
had 1,665,436 paired end sequences with a
mean read length of 149.53 base pairs. The
genome had a mean Phred score of 33.30
and 590 base pairs were dropped after
quality filtering. The SPAdes assembly
contained longer and higher quality contigs
when compared to the Velvet assembly
(Supp. Figure 4) and was used for
downstream analysis and annotation.

The longest contig was 64,486 base pairs and
the total genome was 3,854,901 base pairs
with a 66.63% GC content. The genome was
assembled from 233 contigs and was
determined to have 100% completeness
based on marker lineage across 5,449
genomes (104 markers).

The longest contig was 467,849 base pairs
and the total genome was 13,635,030 base
pairs with a 47.12% GC content. The genome
was assembled from 3,548 contigs and was
determined to have 100% completeness
based on marker lineage across 5,656
genomes (56 markers).

The results from Kbase identified the isolate
Inorganic Farm (98.426 ANI) and Donora 1
(98.4671 ANI) growing in tryptic soy as
closely related to Serratia liquefaciens which
was supported by the BLAST database with
the isolates scoring 99.61% and 99.15%
identity, respectively. Both isolates had
100% query cover.

Kbase results showed that the Donora 3
(89.6983 ANI) isolate in tryptic soy was most
closely related to Bacillus mycoides. After
comparing the 16s sequence to BLAST, it
showed the isolate to match Bacillus cereus,
or a close relative scoring a 97.08% identity
with 100% query coverage.

Donora 2 in tryptic soy contained 1,468,406
paired end sequences with a mean read
length of 149.16 base pairs and a mean
Phred score of 33.31. After preforming
quality filtering, 301 pairs were dropped.
The SPAdes assembly contained longer and
higher quality contigs when compared to the
Velvet assembly (Supp. Figure 3) and was
used for downstream annotation and
analysis.

Pollution 1 isolate growing in tryptic soy
media contained 1,412,215 paired end
sequences and a mean read length of 142.42
base pairs. The mean Phred score was
reported to be 33.31. After preforming
quality filtering 474 base pairs were
dropped. The SPAdes assembly contained
better quality contigs (Supp. Figure 5) and
was used for downstream annotation and
analysis.

The longest contig was 68,093 base pairs and
the total genome was 16,334,954 base pairs
with a 59.1% GC content. The genome was
assembled from 2,345 contigs and
determined to contain 100% completeness
based on marker lineage across 5,656
genomes (56 markers).

The longest contig was 531,839 base pairs
and the total genome was 18,024,254 base
pairs with a 55.91% GC content. The genome
was assembled from 6,810 contigs and
recorded 100% completeness based on

The Kbase results for Donora 2 (98.1358 ANI)
in tryptic soy were inconclusive, however,
the BLAST database confirmed the identity
11

marker lineage across 5,656 genomes (56
markers).

completeness based on marker lineage
across 5,656 genomes (56 markers).

The isolate from Control 2 growing in tryptic
soy media contained 2,214,563 paired end
sequences. There was a mean read length of
140.98 base pairs and a mean Phred score of
33.49. After preforming quality filtering, 527
base pairs were dropped. When compared
to the Velvet assembly, the SPAdes assembly
contained longer and higher quality contigs
(Supp. Figure 6) and was chosen to be used
in downstream annotation and analysis.

The isolate from Control 2 (92.5459 ANI) in
nutrient broth was closely related to
Alcaligenes faecalis and was further
confirmed by BLAST with a percent identity
score of 99.16% and 100% query coverage.
Due to contamination, an additional 16s
genome was identified and ran in BLAST and
confirmed the identity to be Delftia
acidovorans (97.6674 ANI) or a close relative
with a percent identity score of 99.70%.

The longest contig was 390,323 base pairs
and the total genome was 10,278,979 base
pairs with a 55.81% GC content. The genome
was assembled from 1,209 contigs and
contained 100% completeness based on
marker lineage across 5,656 genomes (56
markers).

Isolate Control 2 grown within Lima bean
media contained 1,694,457 paired end
sequences. There was a mean read length of
149.27 base pairs and a mean Phred score of
33.32. The quality filtering resulted in 675
base pairs to be dropped. The SPAdes
assembly contained longer and higher
quality contigs when compared to those of
the Velvet assembly (Supp. Figure 8) and was
chosen to be used in downstream
annotation and analysis.

The Kbase results for the identity of isolates
Pollution 1 (97.9503 ANI) and Control 2 in
tryptic soy were inconclusive, however the
BLAST database suggested that both isolates
were closely related to Alcaligenes faecalis
with 100% query coverage and precent
identity scores of 99.11% and 99.16%
respectively.

The longest contig was 261,853 base pairs
and the total genome was 6,083,381 base
pairs in length with a 58.41% GC content.
The genome was assembled from 2,100
contigs and contained 100% completeness
based on marker lineage across 5,449
genomes (104 markers).

Isolate Control 2 growing in nutrient broth
contained 1,868,298 paired end sequences
with a mean read length of 149.30 base
pairs. The mean Phred score was 33.34 and
after quality filtering 699 base pairs were
dropped. The SPAdes assembly contained
longer and higher quality contigs compared
to the Velvet assembly (Supp. Figure 7) and
was used in downstream analysis and
annotation.

The isolate from Control 2 (91.2187 ANI) in
Lima bean was closely related to
Stenotrophomonas maltophilia as suggested
by Kbase and this was further confirmed by
BLAST scoring a percent identity of 93.54%
and 100% query coverage. During the
genome assembly, it was discovered that all
isolates contained some degree of
contamination (Supp. Table 7). Due to this,
genome assembly was not used for
discussion purposes.

The longest contig was 688,228 base pairs
and the total genome was 16,324,272 base
pairs in length with a 62.55% GC content.
The genome was assembled using 2,833
contigs and determined to have 100%
12

4. Discussion

organic matter is a major source of energy
for microorganisms within the soil and can
stimulate their growth and development.
Changes in the microbial composition will
directly affect organic matter measurements
because the changes in the rate of carbon
and nutrient cycling. Organic content can
reflect the activity of other soil organisms
such as earthworms which are involved in
carbon and nitrogen recycling through
shedding organic residues and promoting
microbial decomposition (Raj & Syriac,
2017).

4.1 Moisture and Organic Content
Both moisture and organic content of soil
directly affect the microbial communities
that reside within it. The ideal agricultural
soil composition is 45% mineral content, 5%
organic matter content, and 50% pore space.
For optimal plant growth, the pore space
should be filled with equal parts water and
air (Blumberg, 1982).
With an optimal moisture content of 25%, all
the samples fell within normal ranges. The
Donora site recorded the closest to optimal
moisture content scoring 23.56%. The
Inorganic orchard recorded a slightly higher
moisture content at 22.06% compared to the
Organic orchard which scored a value of
21.07%. However, the Organic farm scored a
higher moisture content, 23.48%, compared
to the Inorganic farm which recorded a
moisture content of 20% (Figure 4). These
results suggest that the soil isolated from
these sites is relatively healthy in
composition.

Organic matter can work to facilitate heavy
metal release, as well as immobilize it. This
immobilization helps to prevent plant
uptake of toxic metals, decreasing the
pollutant transport and redistribution from
contaminated sites. Organic matter has been
used as a soil amendment in polluted soil for
this reason. Increased soil organic matter
does not reduce the amount of heavy metal
pollutants but reduces the bioavailability of
them to plants. Increase in soil organic
matter has been found to decrease the
amount of cadmium, zinc, and lead in tested
plants (Kwiatkowska-Malina, 2018).

The Control and Pollution sites recorded the
lowest amount of moisture content scoring
19.71% and 19.48% respectively (Figure 4).
Although these scores are lower in
comparison to the other samples, no
conclusions can be drawn from this data that
indicates a substantial difference in soil
composition across the samples. Due to the
similarities observed in moisture content
across all sites, it can be concluded that that
moisture content is not a major contributor
to the structure of microbial communities.

Herbicide application has been found to
increase or decrease the amount of organic
matter present and can be a contributing
factor to abnormally high and low organic
matter scores (Raj & Syriac, 2017). With an
optimal organic matter content of 5%, the
SAI Organic site reported scores closest to
this with the Organic orchard scoring 5.56%
and the Organic farm slightly lower at 4.84%
(Figure 5). These scores combined with the
moisture content scores allow for it to be
concluded that of the sites tested, the SAI
Organic sites contained soil that was the
most agriculturally optimal in composition.

Organic content of soil can be used as an
indicator of soil health because it fluctuates
to adequately reflect biological changes
induced by pollution and contamination.
Although an excess of organic matter is not
favorable to plant growth and development,

The Inorganic site was slightly less optimal
with the Inorganic farm sample recording
13

6.89% and the Inorganic orchard 10.30%
organic matter (Figure 5). This increase in
organic matter could be due to the use of
herbicides which were not present at the SAI
Organic site. Overall, the moisture and
organic content results suggest that the soil
from Inorganic farm was more agriculturally
optimal than that from the Inorganic
orchard, however, both were within normal
ranges suggesting a robust soil microbiome
should be present at both sites.

to drying the samples in the oven, causing an
inaccurate score. Even with this data point
excluded, the soil composition of the Control
site is still less optimal than the Donora, SAI
Organic, and Inorganic sites, but slightly
more than the Pollution site.
It is likely that the agricultural soil samples
were closest to optimal levels due to human
intervention and maintenance. The Control
and Pollution sites lacked the presence of
human
intervention
causing
an
accumulation of organic matter over time,
creating an environment for a diverse
community of microorganisms to thrive on.

The Donora site received a score of 8.95%
organic matter which is comparable to the
optimal score (Figure 5). The organic matter
and moisture content data combined
provides evidence to support that overall,
the soil isolated from the Donora site is in
relatively healthy composition. The higher
than optimal score for organic content could
reflect higher microbial biomass. It has been
found that an increase in organic matter in
soil is indicative of increased enzyme
activity. This suggests that larger microbial
communities are present within the soil
(Bending et al., 2002). The Control and
Pollution sites contained the highest amount
of organic matter scoring 19.71% and
14.84% respectively (Figure 5). These scores
are much higher than the optimal score and
combined with the moisture content data it
can be concluded that the soil collected from
these two sites may contain higher microbial
biomass. The excessive amount of organic
matter could indicate increased enzyme
activity due to the availability of nutrients for
soil microorganisms (Li et al., 2018). A data
point for Control 2 was an extreme outlier
recording a score of 37.35% organic matter,
if this data point is excluded when averaging,
the Control site receives a score of 10.89%
organic matter (Figure 5, Supp. Table 4, &
Supp. Table 5). This data point may have
been caused by failure to remove large rocks
and other debris from the soil sample prior

The Donora, Pollution, and Control sites
elevated levels of organic matter likely
reflect heavy metal contamination within
these soils. The adaption of the soil to
accumulate organic matter would benefit
both plants and soil microorganisms in
polluted areas. An increase in soil organic
matter would immobilize toxic metals,
preventing the reuptake by plants, but also
provide microorganisms with abundant
resources
to
grow
and
develop
(Kwiatkowska-Malina, 2018).
4.2 Zinc Content
A 1998 study was conducted to determine
the heavy metal content of Pennsylvania
soils based on samples collected from the
1980’s. The results state that the mean zinc
content for Washington County was
determined to be 53.44 PPM and 37.29 PPM
for Greene County. The study mentions that
zinc concentrations were higher in surface
horizons than subsurface horizons and that
this is most likely due to man-made
pollutants. The zinc concentrations were
found to be correlated to both organic
matter content and pH among other factors.
(Ciolkosz et al., 1998).

Zinc concentrations were measured at the
Control, Pollution and Donora sites. The
Control site, found within Washington
County contained the lowest amount of zinc
(3.945 PPM) (Table 1). This value is
unexpected due to its proximity to the
Donora site, as well as the reported baseline
zinc content data for that area. It is
suspected that due to the geology of the
Control site, the low amount of zinc could be
attributed to runoff of heavy metals released
during rainfall (Wei et al., 2019).

site were the highest of the experimental
sites, the value still falls within a normal
range (Wade, 2019).
Given the almost 40-year difference
between the soil collections, a change in the
zinc content of the soil is expected, but not
observed unlike in the Control and Pollution
sites where a large drop in zinc
concentration was observed. This data
provides evidence to support the claim that
zinc pollution is persistent across many
decades.

The Pollution site, found within Greene
County contained 21.645 PPM of zinc which
is comparable to the expected value (Table
1). This value is low in comparison to the
data reported data, but the difference in
values could be attributed to the time gap
between studies (Ciolkosz et al., 1998). It is
possible that the difference in the results
and the expected value may be due to the
man-made contribution. Although there is a
pollution source actively affecting this soil, it
may not be emitting the same contaminants.
Power stations have been found to emit
differing heavy metal contaminants such as
sulfur, nitrogen, and carbon oxides, none of
which were examined in this study and may
provide an explanation for the decrease in
zinc as observed by the results (Minnikova et
al., 2017).

4.3 Zinc Growth Curves
Heavy metal contamination of soil
dramatically
effects
the
microbial
communities that live within it. Soil that has
high levels of heavy metal contamination
can substantially reduce the total microbial
biomass as well as significantly alter the
bacterial community structure (Sandaa et
al., 1999).
A 2011 publication was conducted on the
bacteria S. pneumoniae provided evidence
that at concentrations of < 1 mM of zinc
bacterial growth was completely inhibited
(Sandaa et al., 1999). Another study from
1998 notes a decrease in biomass
production at just 0.153 mM of zinc
concentration (Sandaa et al., 1999).
The results from the Lima bean plate (Figure
6A) suggest that the isolate from Control 2
exhibited the most resistance to zinc at 80
mM. The isolate at 80 mM performed better
than all other concentrations, including its
control. This data suggests that the isolate
may contain a metabolic capability that
allows it to utilize zinc at high concentrations
to increase its growth and development. At
this high of a concentration, it was expected
that none of the bacteria would be able to
grow, because this isolate performed better
at the highest concentrations of zinc than

The Donora site, found within Washington
County, contained the highest concentration
of zinc with a score of 51.48 PPM (Table 1).
Although this value is the highest of the sites
examined within this study, it is very similar
to the previously reported values (Ciolkosz
et al., 1998). Although zinc levels differ
greatly across soil (10-300 PPM), there are
many other factors that contribute to zinc
toxicity including soil composition and
organic matter content, among other things.
Although the levels of zinc from the Donora

lower concentrations, it can be suggested
that the isolate has become adapted to a
high zinc environment.

at least 20 mM, or 130.78 PPM, with Donora
1 and Donora 2 isolates (Figure 7A and 7B)
observing inhibition at 40 mM or 261.56
PPM. These results are much higher than the
amount of zinc found within the soil and
again speak to the adaptability of the
isolates.

The results from the nutrient broth plate
(Figure 6B) show that the isolate experience
inhibition at 40 mM. Initial reads indicate
that the isolate thrived in a 5 mM
concentration, performing better than its
control. This data suggests that the bacterial
communities from the Control site contain
zinc resistance capabilities.

The Pollution 1 isolate cultivated in tryptic
soy media (Figure 7D) displayed a more
uniform level of inhibition when compared
to other tested isolates. The isolate appears
to be affected incrementally as the zinc
concentration increases, whereas other
isolates appeared to only be affected
between the 20 mM - 40mM zinc
concentrations (Figure 7). The data from the
zinc content experiment showed that the
soil obtained from the Pollution site
contained 21.645 PPM zinc (Table 1). At 5
mM, or 32.695 PPM, the isolate experienced
very little inhibition. These results were
expected due to the concentration of zinc
found within the soil.

The results from the tryptic soy plate
indicate that the isolate can perform just as
well as its control at the lower three
experimental concentrations of zinc (Figure
6C). These results display the zinc resistant
capabilities of this isolate.
The results for the Control site were
unexpected as it was not anticipated that the
Control site contained high levels of zinc
contamination (Figure 6). These results, in
conjunction with the low zinc levels found at
the site (Table 1), were a surprising
discovery. The findings from the zinc growth
curve experiment suggest that zinc resistant
bacteria found within uncontaminated sites
are still able to persist within high
concentrations of zinc contamination. The
results from the tryptic soy plate (Figure 7AC) show that of the Donora isolates, Donora
3 (Figure 7C) had the lowest amount of zinc
resistant capabilities. All three Donora
isolates cultivated on tryptic soy media
exhibited the ability to grow in 5 mM zinc
with little to no inhibition (Figure 7A-C). Even
at 5 mM zinc, growth was not expected,
therefore these results provide evidence
that the Donora isolates are not inhibited by
relatively high zinc concentrations. The zinc
content test revealed that the soil from the
Donora site contained 51.48 PPM of zinc
(Table 1). The isolates obtained from the
Donora site displayed the capacity to grow in

The isolate from the Inorganic site coined
“IO” experienced minimal inhibition at lower
concentrations of zinc (Figure 7E). Although
zinc content was not measured for the soil
obtained from this site, these results show
that the isolate also retained some degree of
zinc resistance and was comparable to the
capabilities of isolates from other sites.
The data from this experiment revealed the
remarkable capabilities of the bacterial
communities to persist in extremely high
concentrations of zinc. Of the 28 isolates
screened for zinc resistance, 8 exhibited the
ability to grow in elevated levels of zinc
(Figure 6 and 7). Based on the results
received from the zinc content test (Table 1)
it was anticipated that the Donora isolates
would obtain the ability to grow in at least 5
mM or 32.695 PPM, and 10 mM, or 65.39
PPM. Not only did the Donora isolates
16

display this ability, but they also showed very
minimal inhibition at these concentrations
which was an interesting result.

states microbial communities exposed to
long-term zinc pollution will exhibit less
growth
inhibition
in
higher
zinc
concentrations.

Again, based on the results from the zinc
content test (Table 1), it was expected that
the Pollution isolates would be able to grow
in at least 5 mM concentrations of zinc. The
results were consistent with this
expectation, and even displayed the isolate’s
ability to grow within higher concentrations.
Although the isolate experienced a degree of
inhibition at higher concentrations, its ability
to grow at 40 mM, or 261.56 PPM was not
expected and displays zinc resistance for the
isolate.

4.4 Taxonomic Identification of Isolates
Confirming the identity of the isolates that
exhibited resistance to zinc can be extremely
helpful to infer characteristics about the soil
and help to predict the metabolic potential
of the microbial community within it.
Samples TSB Inorganic Farm “IO”, TSB
Donora 1, and TSB Donora 2 were all
identified as Serratia liquefacines. There is
an abundant amount of research confirming
S. liquefacines is commonly found in
industrial effluents known to be polluted
with heavy metals (Kumar et al., 2019;
Ramya & Boominathan, 2017; Zagui et al.,
2021). This species has been reported to
exhibit resistance to cadmium as well as zinc
and has been used in trial studies as an agent
for bioremediation where the bacteria was
isolated from a sample of polluted water and
cultivated in laboratory conditions. The
bacteria displayed a resistance to heavy
metals and was further proven to
significantly
reduce
heavy
metal
concentration by 44.46% when compared to
its control (Kumar et al., 2019).

No growth was expected for the Control
isolates under any concentrations of zinc
based on the results from the zinc content
test (Table 1). The ability of the isolates to
grow at high concentrations of zinc was
surprising and suggests that the isolates
retain zinc resistant capabilities.
Zinc resistance was observed in isolates
across all experimental sites, in both
polluted and unpolluted soils. This data
speaks to the resilience of microorganisms
and to their ability to persist in
contaminated environments.
Of all the sites tested, the Control site is the
closest in proximity to the Donora site and
can be considered soil that may have been
affected by the smog. Due to the low
amount of zinc found within the Control site
soil (3.945 PPM; Table 1), it can be stated
that the isolates were not forced to exist
within a zinc contaminated environment.
The proximity to the pollution event can help
to explain the zinc resistance capabilities of
these isolates in the Lima bean and nutrient
broth plates. Due to this, the zinc resistance
observed in the Control isolates
corroborates the initial hypothesis that

A study conducted to investigate the
capabilities
of
Serratia
liquefacines
identified a total of 110 genes related to iron
acquisition and metabolism. Of these genes,
49 were associated with siderophore
biosynthesis,
secretion,
and
iron
internalization. The other 61 genes were
associated with metal metabolism. It was
determined through experimentation that a
specific genome within the Serratia
liquefacines clade, SlFG3, produces two
siderophores, one of which has chemical
characteristics of catecholates that produces

a purple color after development. This stain
was also found in extreme environmental
conditions and displayed high adaptability
(Caneschi et al., 2019). This research helps to
explain the purple hue that was emitted
from the TSB Inorganic farm colony and
displays the unique capabilities and
adaptability of this strain of bacteria.

copper, zinc, lead, and chromium. A strain of
Alcaligenes faecalis, phenolicus MB207
underwent genome sequencing and
annotation to document the genes found
within it and the capabilities they contain.
Genes commonly used in metal transport
inside and outside of the cell were detected
within the analyzed genome commonly used
in metal detoxification and survival in high
metal stressed environments. The genome
has also been associated with nanoparticle
production as well as the conversion of the
most toxic form of arsenic to its less
dangerous form (Basharat et al., 2018). The
results from the zinc growth curves show
differing inhibitions for these isolates, where
Control 2 LB showed preference to the 80
mM concentration, performing better than
the control (Figure 6). Pollution 1, however,
did not show this preference to this
concentration (Figure 7). This suggests that
these isolates may be differing strains of
Alcaligenes faecalis. This species is another
contender for use in bioremediation due to
the zinc resistant capabilities it possesses.
Although it contains metal detoxifying
properties, this bacterium is known to be a
common soil dweller and so its presence was
not unexpected. Due to it being ubiquitous,
it is expected to be found in all
environments, polluted and nonpolluted.

While all three isolates were determined to
be Serratia liquefaciens, only one of the
three isolates exhibited purple growth, the
Inorganic Farm isolate. It is also important to
note that the three isolates displayed
differing degrees of zinc resistance (Figure
7). Isolate Donora 1 displayed a higher
degree of inhibition when compared to the
other two isolates. While all isolates
obtained the ability to grow at 40 mM
concentration, isolate Donora 1 performed
better when compared to its control than
the other two isolates (Figure 7). These
findings suggest that the isolates appear to
be different strains of Serratia liquefaciens.
The TSB Donora 3 bacteria was determined
to be Bacillus cereus, a common soil
bacterium, which has been identified as
having metal resistance capabilities to lead,
cadmium, and chromium. A study conducted
on the NWUAB01 strain of Bacillus cereus
discovered the abundance of heavy metal
resistant genes against arsenic, cadmium,
copper, cobalt, and zinc within the genome.
These defining characteristics of this species
make it another promising contender for use
in bioremediation techniques and help to
explain its presence in areas effected by
heavy
metal
industrial
pollution
(Avangbenro & Babalola, 2020).

NB Control 2 contained two hits from BLAST
with a high match, therefore the identity of
this isolate could not be determined. The top
hits were Alcaligenes faecalis and Delftia
acidovorans.
As
mentioned
above,
Alcaligenes faecalis has been previously
shown to have metal resistance capabilities.
The Delftia genus has also been found to be
resistant to zinc, lead, selenium, copper,
aluminum, and nickel. A specific Delftia
species has also been proved to have lead
and zinc sorption capacities (BautistaHernandez et al., 2012). These findings are

The bacteria from TSB Pollution 1 and TSB
Control 2 were identified to be Alcaligenes
faecalis which has been reported to
demonstrate tolerance to heavy metal
micropollutants such as nickel, cadmium,
18

supported by the zinc growth curve obtained
in this study which showed that the isolate
at 80 mM concentration performed better
when compared to 40 mM concentration.

community structure, where they may be
less
prevalent
in
uncontaminated
environments. Further research involving
the overall microbial community structure in
contaminated
and
uncontaminated
environments could provide an insight into
the competitiveness of these bacteria and
their persistence.

The LB Control 2 sample was positively
identified to be Stenotrophomonas
maltophilia, a heavily studied soil bacterium
with remarkable abilities. It has been
classified as an opportunistic pathogen that
causes nosocomial infections and has also
displayed antibiotic resistance (Sanchez,
2015). S. maltophilia has also been shown to
tolerate high levels of toxic metals including
cadmium, lead, copper, zinc, mercury, silver,
selenite, tellurite, and uranyl (Pages et al.,
2008). A 2008 study provides data that
suggests that in addition to the high
tolerance of antibiotics, the bacterium has
developed at least two different
mechanisms to overcome metal toxicity and
detoxification of cadmium to cadmium
sulfide (Pages et al., 2008).

5. Conclusion
The findings of this study suggest that
industrial pollution can prevail for decades
following the pollution event. The bacteria
isolated from the soil has displayed the
capabilities of growing in heavy metal
contaminated environments, especially zinc,
and suggest that this characteristic has aided
in their survival within the polluted soil. The
soil from Donora contained the highest zinc
content, providing evidence of persisting
pollution.
The results from this study show that the
damage caused by the Donora Smog event
and the long years of industrialization in
Donora still effects the soil and the
microorganisms within it to this day.

While zinc resistant genes within the
isolates’ genome could not be conclusively
confirmed due to contamination, previously
published literature supports the ability of
the identified organisms to confer zinc
resistance. The ability of all of species
identified to be resistant to heavy metals is
synonymous with the expected results. Zinc
resistance was found across varying soil
conditions,
not
only
contaminated
environments. This suggests that these
microorganisms do not depend on the
contamination for survival, but rather persist
within a wide range of environments,
utilizing their zinc resistant capabilities
should
that
environment
become
contaminated. This ability to survive within
polluted soils would allow them to out
compete
more
metal
sensitive
microorganisms within contaminated sites,
like Donora, shifting the microbial

6. References
1948 smog. Donora Historical Society and Smog Museum.
(n.d.). Retrieved April 28, 2022, from
https://www.sites.google.com/site/donorahistorica
lsociety/donora-history/1948-smog
Ayangbenro, A. S., & Babalola, O. O. (2020). Genomic
analysis of bacillus cereus NWUAB01 and its heavy
metal removal from polluted soil. Scientific Reports,
10(1). https://doi.org/10.1038/s41598-020-75170-x
Balafrej, H., Bogusz, D., Triqui, Z.-E. A., Guedira, A., Bendaou,
N., Smouni, A., & Fahr, M. (2020). Zinc
hyperaccumulation in plants: A Review. Plants, 9(5),
1–22. https://doi.org/10.3390/plants9050562
Baranauskas, L. (2017, November 29). The historically hazy
story of Donora's deadly smog. Atlas Obscura.
Retrieved April 25, 2022, from
https://www.atlasobscura.com/articles/donorasmog-1948

Basharat, Z., Yasmin, A., He, T., & Tong, Y. (2018). Genome
sequencing and analysis of alcaligenes faecalis
subsp. phenolicus MB207. Scientific Reports, 8(1).
https://doi.org/10.1038/s41598-018-21919-4

Ivel, J. (n.d.). Donora, Pennsylvania Smog Event of 1948.
Donora, pennsylvania smog event of 1948.
Retrieved April 25, 2022, from
http://www.soe.uoguelph.ca/webfiles/gej/AQ2017
/Ivel/index.html

Bautista-Hernández, D. A., Ramírez-Burgos, L. I., DuranPáramo, E., & Fernández-Linares, L. (2012). Zinc and
lead Biosorption by Delftia tsuruhatensis: A
bacterial strain resistant to metals isolated from
mine tailings. Journal of Water Resource and
Protection, 04(04), 207–216.
https://doi.org/10.4236/jwarp.2012.44023

Jacobs, E. T., Burgess, J. L., & Abbott, M. B. (2018). The
Donora Smog Revisited: 70 years after the event
that inspired the Clean Air Act. American Journal of
Public Health, 108(S2).
https://doi.org/10.2105/ajph.2017.304219
Kour, R., Bhojiya, A. A., Meena, R. H., Singh, A., Mohanty, S.
R., Rajpurohit, D., & Ameta, K. D. (2020). Zinc
tolerant plant growth promoting bacteria alleviates
phytotoxic effects of zinc on maize through zinc
immobilization. Scientific Reports, 10(1), 1–13.
https://doi.org/10.1038/s41598-020-70846-w

Bending, G. D., Turner, M. K., & Jones, J. E. (2002).
Interactions between crop residue and soil organic
matter quality and the functional diversity of soil
microbial communities. Soil Biology and
Biochemistry, 34(8), 1073–1082.
https://doi.org/10.1016/s0038-0717(02)00040-8

Kour, R., Jain, D., Bhojiya, A. A., Sukhwal, A., Sanadhya, S.,
Saheewala, H., Jat, G., Singh, A., & Mohanty, S. R.
(2019). Zinc biosorption, biochemical and molecular
characterization of plant growth-promoting zinctolerant bacteria. 3 Biotech, 9(11), 1–17.
https://doi.org/10.1007/s13205-019-1959-2

Bleiwas, D. I., & DiFrancesco, C. (2010). Historical zinc
smelting in New Jersey, Pennsylvania, Virginia, West
Virginia, and Washington, D.C., with estimates of
atmospheric zinc emissions and other materials.
U.S. Geological Survey Open File Report, 1131.
https://doi.org/10.3133/ofr20101131

Kumar, P., Gupta, B. S., Anurag, & Soni, R. (2019).
Bioremediation of Cadmium by Mixed Indigenous
Isolates Serratia liquefaciens BSWC3 and Klebsiella
Pneumoniae RpSWC3 Isolated from Industrial and
Mining Affected Water Samples. Pollution, 5(2),
351–360.
https://doi.org/10.22059/poll.2018.268603.533

Blumberg, B. (1982). An Introduction to Soils of Pennsylvania
(dissertation). The Pennsylvania State University,
University Park, PA.
Caneschi, W. L., Sanchez, A. B., Felestrino, É. B., Lemes, C. G.,
Cordeiro, I. F., Fonseca, N. P., Villa, M. M., Vieira, I.
T., Moraes, L. Â., Assis, R. de, do Carmo, F. F.,
Kamino, L. H., Silva, R. S., Ferro, J. A., Ferro, M. I.,
Ferreira, R. M., Santos, V. L., Silva, U. de, Almeida,
N. F., … Moreira, L. M. (2019). Serratia liquefaciens
FG3 isolated from a metallophyte plant sheds light
on the evolution and mechanisms of adaptive traits
in extreme environments. Scientific Reports, 9(1),
1–16. https://doi.org/10.1038/s41598-019-54601-4

Kwiatkowska-Malina, J. (2018). Functions of organic matter
in polluted soils: The effect of organic amendments
on phytoavailability of heavy metals. Applied Soil
Ecology, 123, 542–545.
https://doi.org/10.1016/j.apsoil.2017.06.021
Li, L., Xu, M., Eyakub Ali, M., Zhang, W., Duan, Y., & Li, D.
(2018). Factors affecting soil microbial biomass and
functional diversity with the application of organic
amendments in three contrasting cropland soils
during a field experiment. PLOS ONE, 13(9).
https://doi.org/10.1371/journal.pone.0203812

Ciolkosz, E. J., Stehouwer, R. C., & Amistadi, M. K. (1998).
Metals Data for Pennsylvania Soil (dissertation).
Pennsylvania State University, University Park, PA.
Environmental Pollution Centers. (2022). Zinc poisoning.
Environmental Pollution Centers. Retrieved April
25, 2022, from
https://www.environmentalpollutioncenters.org/zi
nc/

Minnikova, T. V., Denisova, T. V., Mandzhieva, S. S.,
Kolesnikov, S. I., Minkina, T. M., Chaplygin, V. A.,
Burachevskaya, M. V., Sushkova, S. N., & Bauer, T.
V. (2017). Assessing the effect of heavy metals from
the Novocherkassk power station emissions on the
biological activity of soils in the adjacent areas.
Journal of Geochemical Exploration, 174, 70–78.
https://doi.org/10.1016/j.gexplo.2016.06.007

Greenspec. (2022). Zinc Production & Environmental Impact.
Greenspec. Retrieved April 25, 2022, from
https://www.greenspec.co.uk/building-design/zincproduction-environmental-impact/

Pages, D., Rose, J., Conrod, S., Cuine, S., Carrier, P., Heulin,
T., & Achouak, W. (2008). Heavy Metal Tolerance in
Stenotrophomonas Maltophilia. PLoS ONE, 3(2).
https://doi.org/10.1371/journal.pone.0001539

Helfand, W. H., Lazarus, J., & Theerman, P. (2001). Donora,
Pennsylvania: An environmental disaster of the
20th Century. American Journal of Public Health,
91(4), 553–553.
https://doi.org/10.2105/ajph.91.4.553

Raj, S. K., & Syriac, E. K. (2017). Herbicidal effect on the bioindicators of soil health- A Review. Journal of
Applied and Natural Science, 9(4), 2438–2448.
https://doi.org/10.31018/jans.v9i4.1551

Maltophilia. Frontiers in Microbiology, 6.
https://doi.org/10.3389/fmicb.2015.00658
Wade, K. M. (2019). Zinc. plantprobs.net. Retrieved April 28,
2022, from
https://plantprobs.net/plant/nutrientImbalances/zi
nc.html

Ramya, R., & Boominathan, M. (2017). Isolation of Serratia
Liquefaciens as Metal Resistant Bacteria from
Industrial Effluent. International Journal of Advance
Research, Ideas and Innovations in Technology, 3(6),
1272–1275.

Wei, L., Liu, Y., Routh, J., Tang, J., Liu, G., Liu, L., Luo, D., Li,
H., & Zhang, H. (2019). Release of heavy metals and
metalloids from two contaminated soils to surface
runoff in southern China: A simulated-rainfall
experiment. Water, 11(7), 1339.
https://doi.org/10.3390/w11071339

Rossi, R. J., Bain, D. J., Hillman, A. L., Pompeani, D. P.,
Finkenbinder, M. S., & Abbott, M. B. (2017).
Reconstructing early industrial contributions to
legacy trace metal contamination in southwestern
Pennsylvania. Environmental Science & Technology,
51(8), 4173–4181.
https://doi.org/10.1021/acs.est.6b03372

Zagui, G. S., Moreira, N. C., Santos, D. V., Darini, A. L.,
Domingo, J. L., Segura-Muñoz, S. I., & Andrade, L. N.
(2021). High occurrence of heavy metal tolerance
genes in bacteria isolated from wastewater: A new
concern? Environmental Research, 196, 110352.
https://doi.org/10.1016/j.envres.2020.110352

Sandaa, R.-A., Torsvik, V., Enger, Ã. I., Daae, F. L., Castberg,
T., & Hahn, D. (1999). Analysis of bacterial
communities in heavy metal-contaminated soils at
different levels of resolution. FEMS Microbiology
Ecology, 30(3), 237–251.
https://doi.org/10.1111/j.15746941.1999.tb00652.x
Snyder, L. P. (1994). “The death-dealing smog over Donora,
Pennsylvania”: Industrial Air Pollution, public health
policy, and the politics of expertise, 1948–1949.
Environmental History Review, 18(1), 117–139.
https://doi.org/10.2307/3984747
Sánchez, M. B. (2015). Antibiotic resistance in the
opportunistic pathogen Stenotrophomonas

APPENDICES
A.1 Soil Collection Data
Supplemental Table 1. Soil Collection Data for the Seven Sample Sites
Sample Name
Donora
Control
Pollution
Organic Farm
Organic Orchard
Inorganic Farm
Inorganic Orchard

Collection
Date
05/17/2021
05/24/2021
05/17/2021
05/17/2021
05/17/2021
05/20/2021
05/20/2021

GPS Coordinate

Address

(40°10’41” N 79°51’11” W)
(40°02’40” N 79°54’51” W)
(39°52’44” N 79°55’13” W)
(40°02’50” N 79°54’49” W)
(40°02’50” N 79°54’49” W)
(40°04’48” N 79°08’20” W)
(40°04’48” N 79°08’20” W)

470 Galiffa Dr. Donora PA
377 E Malden Dr. Coal Center PA
41 Haig Ave. Carmichaels PA
377 E Malden Dr. Coal Center PA
377 E Malden Dr. Coal Center PA
1665 Coxes Creek Rd. Somerset PA
745 Edie Rd. Somerset PA

A.2 Soil Moisture Raw Data
Supplemental Table 2: Raw Data Moisture Content Test
Sample Name

Inorganic Orchard 1
Inorganic Orchard 2
Inorganic Orchard 3
Inorganic Farm 1
Inorganic Farm 2
Inorganic Farm 3
Organic Farm 1
Organic Farm 2
Organic Farm 3
Pollution 1
Pollution 2
Pollution 3
Organic Orchard 1
Organic Orchard 2
Organic Orchard 3
Control 1
Control 2
Control 3
Donora 1
Donora 2
Donora 3

Weight
of the
foil
2.01
2.02
2.02
2.01
2.03
1.98
1.99
2
2.02
2
2.02
2.02
2.02
2
2.02
2.02
2
2.02
2.01
2.04
2.01

Wet Weight
of soil
17.68
10.44
12.54
22.25
25.87
26.85
27.68
23.64
26.12
10.02
13.48
15.22
21.42
24.76
21.01
14.43
8.95
13.81
10.42
18.22
16.97

Dry Weight
of soil with
foil
16.16
10.07
11.64
20.67
21.85
23.33
22.58
20.54
22.06
9.58
13.37
14.46
18.28
22.08
18.78
13.34
8.57
14.31
9.73
16.16
15.2

Dry weight
of soil only

Calculated
Moisture %

14.15
8.05
9.62
18.66
19.82
21.35
20.59
18.54
20.04
7.58
11.35
12.44
16.26
20.08
16.76
11.32
6.57
12.29
7.72
14.12
13.19

19.97
22.89
23.29
16.13
23.39
20.48
25.61
21.57
23.27
24.35
15.81
18.27
24.09
18.9
20.23
21.55
26.59
11.01
25.91
22.5
22.27

Supplemental Table 3: Average Moisture Content Test and Calculated Standard Deviation
Site Name
Inorganic Orchard
Inorganic Farm
Organic Farm
Pollution
Organic Orchard
Control
Donora

Average
Moisture %
22.06
20.00
23.48
19.48
21.07
19.71
23.56

Standard
Deviation
1.789450567
3.653724128
2.028431249
4.396013345
2.695817749
7.950153038
2.03840624

A.3 Soil Organic Matter Raw Data
Supplemental Table 4: Raw Data Organic Matter Content Test
Sample Name

Weight
of
crucible
7.18
17.56
17.51
29.46
17.58
29.25
16.62
16.68
30.48
27.37
16.58
15.96
16.54
17.16
16.68
28.62
9.44
10.58
6.57
10.23
7.28

Weight
of soil
5.31
5.57
5.76
5.15
5.24
5.05
5.81
5.69
5.63
5.22
5.27
5.44
5.6
5.83
5.28
5.59
5.06
5.23
5.11
5.3
5.25

Dry Weight of
soil and
crucible
12.05
22.46
22.66
33.96
22.44
34.26
22.09
22.17
35.82
31.86
20.88
20.74
21.85
22.64
21.67
33.42
12.61
15.41
11.15
15.02
12.17

Dry weight of
soil only
4.87
4.9
5.15
4.5
4.86
5.01
5.47
5.49
5.34
4.49
4.3
4.78
5.31
5.48
4.99
4.8
3.17
4.83
4.58
4.79
4.89

Calculated
Organic Matter
%
8.29
12.03
10.59
12.62
7.25
0.79
5.85
3.51
5.15
13.98
18.41
12.13
5.18
6.00
5.49
14.13
37.35
7.65
10.37
9.62
6.86

Supplemental Table 5: Average Organic Matter Content Test and Calculated Standard
Deviation
Site Name
Inorganic Orchard
Inorganic Farm
Organic Farm
Pollution
Organic Orchard
Control
Donora

Average Organic
Matter %
10.30
6.89
4.84
14.84
5.56
19.71
8.95

Standard
Deviation
1.8878457
5.922661268
1.199290492
3.223336909
0.416337915
15.61777227
1.851225103

A.4 Zinc Content Raw Data
Supplemental Table 6: Absorbency readings and calculated zinc concentrations
Site
Name
Control
Donora
Pollution

Absorbency Concentration Calculated Zinc
Content (PPM)*
-0.0086
0.0789
3.945
0.087
1.0295
51.48
0.027
0.4329
21.645

*The calculated average absorbency of isolates from Pollution, Control, and Donora sites after 9 reads.

A.5 Growth Curve Raw Data
Supplemental Table 7: Absorbency readings Control 2 LB Results

Supplemental Table 8: Absorbency readings Control 2 NB Results

Supplemental Table 9: Absorbency readings Control 2 TSB Results

Supplemental Table 10: Absorbency readings Donora 1 TSB Results

Supplemental Table 11: Absorbency readings Donora 2 TSB Results

Supplemental Table 12: Absorbency readings Donora 3 TSB Results

Supplemental Table 13: Absorbency readings Pollution 1 TSB Results

Supplemental Table 14: Absorbency readings Inorganic Farm TSB Results

A.5 Genome Assembly Data
Supplemental Table 7: Shotgun Sequencing Data Assembly Results
Total
Sequences
Inorganic Farm
TSB

1,984,094

Mean
read
length
147.68

Donora 1 - TSB
Donora 2 - TSB
Donora 3 - TSB
Pollution 1 - TSB
Control 2- TSB
Control 2 – NB
Control 2 - LB

1,759,608
1,468,406
1,665,436
1,412,215
2,214,563
1,868,298
1,694,457

146.44
149.16
149.53
142.42
140.98
149.30
149.27

Phred
score

Completeness

Contamination

ANI

33.36

100%

258.71

98.426

33.38
33.31
33.30
33.31
33.49
33.34
33.32

100%
100%
100%
100%
100%
100%
100%

243.76
206.13
145.47
265.16
133.26
237.97
34.96

98.4671
98.1358
89.6983
97.9503
98.0441
92.5459
91.2187

Supplemental Figure 1: Inorganic Farm TSB QUAST Comparison

Supplemental Figure 2: Donora 1 TSB QUAST Comparison

Supplemental Figure 3: Donora 2 TSB QUAST Comparison

Supplemental Figure 4: Donora 3 TSB QUAST Comparison

Supplemental Figure 5: Pollution 1 TSB QUAST Comparison

Supplemental Figure 6: Control 2 TSB QUAST Comparison

Supplemental Figure 7: Control 2 NB QUAST Comparison

Supplemental Figure 8: Control 2 LB QUAST Comparison