Journal Article Review.

RESEARCH ARTICLE

The environmental risk assessment of cell-

processing facilities for cell therapy in a

Japanese academic institution

Mitsuru Mizuno 1,2

, Kentaro Endo 1 , Hisako Katano

1,2 , Ayako Tsuji

2 , Naomi Kojima

1 ,

Ken Watanabe 1 , Norio Shimizu

1,2 , Tomohiro Morio

2,3 , Ichiro SekiyaID

1,2*

1 Center for Stem Cell and Regenerative Medicine, Tokyo Medical and Dental University, Bunkyo-ku,

Yushima, Tokyo, Japan, 2 Center for Transfusion Medicine and Cell Therapy, Tokyo Medical and Dental

University, Bunkyo-ku, Yushima, Tokyo, Japan, 3 Department of Pediatrics and Developmental Biology,

Graduate School, Tokyo Medical and Dental University, Bunkyo-ku, Yushima, Tokyo, Japan

* [email protected]

Abstract

Cell therapy is a promising treatment. One of the key aspects of cell processing products is

ensuring sterility of cell-processing facilities (CPFs). The objective of this study was to

assess the environmental risk factors inside and outside CPFs. We monitored the tempera-

ture, humidity, particle number, colony number of microorganisms, bacteria, fungi, and

harmful insects in and around our CPF monthly over one year. The temperature in the CPF

was constant but the humidity fluctuated depending on the humidity outside. The particle

number correlated with the number of entries to the room. Except for winter, colonies of

microorganisms and harmful insects were detected depending on the cleanliness of the

room. Seven bacterial and two fungal species were identified by PCR analyses. Psocoptera

and Acari each accounted for 41% of the total trapped insects. These results provide useful

data for taking the appropriate steps to keep entire CPFs clean.

Introduction

Regenerative medicine using stem cells holds great promise for the treatment of some incur-

able diseases. In particular, mesenchymal stem cells (MSCs) [1] and induced pluripotent stem

cells (iPSCs) [2] have been considered feasible cell sources. In recent years, many researchers

have conducted clinical trials for developing new cell products [3–5]. In Japan, cell products

are processed at each academic cell-processing facility (CPF) based on a new regulatory frame-

work for regenerative medicine [6, 7]. The safety of cell products is threatened not only by

undesired cell characteristics, such as tumorigenicity, but also by cell culture contamination

caused by microorganisms. Sterilization is the simplest way to eliminate such infections; how-

ever, living cell products cannot be sterilized. Therefore, the most important aspect of cell pro-

cessing safety is guaranteeing the sterility of the process.

The sterility of cell products is secured by the appropriate management of CPF environ-

ments and traceability based on these records. Such environmental management policy should

PLOS ONE

PLOS ONE | https://doi.org/10.1371/journal.pone.0236600 August 5, 2020 1 / 14

a1111111111

a1111111111

a1111111111

a1111111111

a1111111111

OPEN ACCESS

Citation: Mizuno M, Endo K, Katano H, Tsuji A,

Kojima N, Watanabe K, et al. (2020) The

environmental risk assessment of cell-processing

facilities for cell therapy in a Japanese academic

institution. PLoS ONE 15(8): e0236600. https://doi.

org/10.1371/journal.pone.0236600

Editor: Raghvendra Bohara, National University of

Ireland, Galway, IRELAND

Received: May 20, 2020

Accepted: July 8, 2020

Published: August 5, 2020

Peer Review History: PLOS recognizes the

benefits of transparency in the peer review

process; therefore, we enable the publication of

all of the content of peer review and author

responses alongside final, published articles. The

editorial history of this article is available here:

https://doi.org/10.1371/journal.pone.0236600

Copyright: © 2020 Mizuno et al. This is an open access article distributed under the terms of the

Creative Commons Attribution License, which

permits unrestricted use, distribution, and

reproduction in any medium, provided the original

author and source are credited.

Data Availability Statement: All relevant data are

within the paper and its Supporting Information

files.

be planned to depend on CPF structure to maintain proper cleanliness and eliminate the risk

of extrinsic contamination of the process [8, 9]. The risk of extrinsic contamination can be

affected by the location of the CPF, the environment inside or outside the CPF, and the degree

of human intervention during the process [10, 11]. These environmental risk factors include

temperature, humidity, number of particles, entrance time to a clean room, airborne microor-

ganisms, and harmful insects. In Japan, the temperature and humidity fluctuate according to

the season. Therefore, to perform an appropriate environmental risk assessment, it is essential

to obtain total environmental data throughout the year. Since environmental factors may

interact each other, the spatiotemporal relationships among them should be also evaluated.

The objective of this study was to assess the environmental risk factors inside and outside a

CPF to guarantee the safety of cell products. We collected comprehensive environmental data

throughout the year during an ongoing clinical study and investigated the relationships among

them. Their effects on the sterility of processed cell products were also assessed.

Material & methods

Classification of areas in the cell-processing facility (CPF)

The CPF was divided into uncontrolled and controlled areas (Fig 1A). The uncontrolled area

was defined as the general environment (GE) and the controlled areas were divided into four

areas (Grade D, Grade C, Grade B, and Grade A) according to cleanliness defined by “consid-

eration on aseptic operation in cell culture processing facilities” based on the “Safety Act” pub-

lished by the Japanese Society for Regenerative Medicine (S1 Table). Grade A was an area

where aseptic operation was performed.

Fig 1B shows the schematic diagram of our CPF. Operators entered from the entrance,

went through the first gowning room, the buffer room, and the second gowning room in that

order, entered the cell-processing room, and worked using a safety cabinet. The operators then

passed through the second de-gowning room, the buffer room, and the first de-gowning room

before exiting.

In this study, in front of the facility, the corridor and workspace outside the CPF, and the

laboratory outside the CPF were considered the GE. The entrance of the CPF was labeled

Grade D; the first gowning room, the buffer room, and the supply room were labeled Grade C;

the second gowning room and cell-processing room were labeled Grade B; and the space in

the safety cabinet was labeled Grade A.

Data collection

Assessment items were collected from January 2018 to December 2018. The data in the CPF

during a certain period in February and March were not collected due to calibration and vali-

dation of the CPF for an annual inspection. The total number of entrance times in a month

was counted based on the cell-processing record.

Monitoring of temperature and humidity

The temperature and humidity in each area were measured using the precision hair hygro-

thermometer (Higest II: Sato Keiryoki Mfg. Co., Ltd., Tokyo, Japan). The temperature outside

the building was based on the database of the Japan Meteorological Agency. The mean values

of temperature and humidity in the Grade D, Grade C, and Grade B areas were calculated

from data automatically measured every 10 minutes (TH-EV6A, Shinyei Technology Co., Ltd.,

Kobe, Japan).

PLOS ONE The environmental risk assessment of cell-processing facilities

PLOS ONE | https://doi.org/10.1371/journal.pone.0236600 August 5, 2020 2 / 14

Funding: The Research Project for Practical

Applications of Regenerative Medicine

(JP18bk0104065) supported IS from the Japan

Agency for Medical Research and Development

(AMED) to complete this study.

Competing interests: The authors have declared

that no competing interests exist.

Monitoring of particles

The number of particles (� 0.5 μm) per m3 was manually obtained using METONE HHPC 3+ (Beckman Coulter, Inc., CA, USA) every two weeks. The maximum value was used. The inte-

grated number of particles (� 5μm) was also obtained only in Grade B every 10 minutes using an automatic device (KA-02, Rion Co., Ltd., Tokyo, Japan). The data in cleanliness-controlled

areas were shown as the maximum value in each room. Obtained data were divided in opera-

tion (during processing) and at rest (not during operation). Pearson’s correlation coefficient

between the number of entrance times to Grade B and the integrated number of particles in

Grade B at all times including in operation and at rest was further analyzed.

Monitoring of microorganisms

One thousand L of air was aspirated by Biosamp (Midori Anzen Co., Ltd., Tokyo, Japan), and

airborne microorganisms were cultured on soybean-casein digest agar (Becton, Dickinson and

Company, NJ, USA). After five days of culture, macroscopic images of colonies were captured

by M165 FC (Leica Microsystems GmbH, Wetzlar, Germany). In addition, settle plate and

glove print tests were performed using a soybean-casein digest agar to assess the presence of

microorganisms in Grade A. The quantification of colonies was visually evaluated after five

Fig 1. Assessment items for each room of the Cell-Processing Facility (CPF). (a) Classification of areas in the CPF according to cleanliness defined by the Japanese

Society for Regenerative Medicine (S1 Table). Assessed area is divided into uncontrolled and controlled areas. The uncontrolled area is defined as the general

environment (GE), and the controlled areas are divided into four areas according to cleanliness. (b) Schematic diagram of CPF and assessment items in each room.

Operators enter from the entrance, go through the first gowning room, the buffer room, and the second gowning room in that order, enter the cell-processing room,

and work using a safety cabinet. The operators then pass through the de-gowning room, the buffer room, and the first gowning room before exiting. The colors

represent the cleanliness grades. The illustrations show assessment items for #1, “temperature and humidity”; #2, “particle count”; #3, “microbial monitoring”; and #4,

“insect” performed in each room. Insects were also monitored in the corridor and workspace, laboratory, and in front of the CPF.

https://doi.org/10.1371/journal.pone.0236600.g001

PLOS ONE The environmental risk assessment of cell-processing facilities

PLOS ONE | https://doi.org/10.1371/journal.pone.0236600 August 5, 2020 3 / 14

days. The data in cleanliness-controlled areas were shown as the maximum value in each

room. The chi-square test was performed to assess the association between humidity and the

number of airborne microorganisms in Grade C and Grade D (test number = 133). An odds

ratio was calculated for the rate of the occurrence of airborne microorganisms.

Identification of bacterial and fungus species

Colonies of microorganisms were manually picked up and their nucleic acid extracted using a

DNA extracting kit (QIAamp DNA mini kit, Qiagen, Venlo, Netherlands). Nucleic acids were

amplified by PCR using a bacterial ribosomal RNA-specific designed primer and probe set

[12]. Real-time PCR was performed in a LightCycler 480 instrument (Roche Diagnostics,

Basel, Switzerland). For PCR reaction, the following reagents were used: forward primer; agg-

cagcagtDRggaat reverse primer; ggactacYVgggtatctaat probe FAM-tgccagcagccgcggtaatacR-

Dag-BHQ and the LightCycler 480 Probes Master (Roche Diagnostics). DRYV in the sequence

of primer and probe represents mixed bases defined by the International Union of Pure and

Applied Chemistry. The PCR products were directly sequenced with Applied Biosystems

3130xl Genetic Analyzer (Thermo Fisher Scientific, Waltham, MA, USA). Bacterial and fungus

species were identified by querying NCBI Nucleotide BLAST sequence data [13]. The species

suggested by BLAST with an identified rate over 99% was considered a candidate species.

Monitoring of harmful insects

Traps (Earth Environmental Service Co., Ltd., Osaka, Japan) were settled in each area. After

31 ± 7 days, the number of insects per trap was counted, and the species were identified macro- scopically or using a stereo microscope. The number of insects per trap in the corridor outside

the CPF was compared below 15˚C and 15˚C or more outside the building by the Mann Whit-

ney U-test.

Cell processing and sterility test

For clinical research “intraarticular injections of synovial stem cells for osteoarthritis of the

knee” (UMIN000026732), cell products were prepared in the CPF. Detailed processing proto-

cols have been reported previously [4]. Briefly, synovium was treated with enzyme and cul-

tured in a medium containing autologous serum and antibiotics for 14 days. Twenty million

cells were administered intraarticularly. According to the regulations of Japanese Pharmaco-

poeia, part of the cell products was subjected to a sterility test using fluid thioglycolate medium

(Becton, Dickinson and Company) and soybean-casein digest broth (Becton, Dickinson and

Company). These media were incubated at 22.5 ± 2.5˚C and 32.5 ± 2.5˚C, respectively. After 14 days, the presence or absence of microorganisms was visually evaluated.

Ethics approval and consent to participate

This clinical study (RM2018-006) was approved by the certified special committee for regener-

ative medicine in Tokyo Medical and Dental University (committee reference number:

NA8140003) and submits a plan to the Minister of Health, Labor, and Welfare (MHLW,

PB3160032). This clinical study was carried out in accordance with the Helsinki Declaration.

Statistical analysis

All statistical analysis was performed with Prism 8 software (GraphPad Inc., La Jolla, CA,

USA). Each statistical analysis is described in the material and methods section and figure leg-

ends. Two-tailed p values smaller than 0.05 were considered statistically significant.

PLOS ONE The environmental risk assessment of cell-processing facilities

PLOS ONE | https://doi.org/10.1371/journal.pone.0236600 August 5, 2020 4 / 14

Results

Monitoring of temperature, humidity, and number of particles

Although the temperature outside the building changed with seasons, that in the CPF was con-

stant throughout the year (Fig 2). The temperature in Grade B and Grade C was always a few

degrees lower than that in Grade D. The humidity in Grade B–D varied largely depending on

that outside the building with highs in summer and lows in winter.

Monitoring of particles

Monthly transitions in the number of particles (� 0.5μm) differed from Grade B–D (Fig 3A). In Grade A, the particle number was almost zero except for February, when it was not measured. The

number of particles was well below the acceptable limit, which was under 3,520 for Grade A and B,

35,200 for Grade C, and 3.52 million for Grade D. The integrated number of particles (� 0.5μm) in Grade B correlated with the integrated number of entries to the cell-processing room (Fig 3B,

R = 0.68, p = 0.03). The integrated number of particles (� 5μm) in Grade B was also correlated with the integrated number of entries to the cell-processing room (S1 Fig, R = 0.64, p = 0.048).

Monitoring of microorganisms and identification of species

Monthly transitions of the colony number of airborne microorganisms differed from Grade

B–D (Fig 4A). In Grade A, the colony number was completely zero except for February, when

Fig 2. Monthly transitions of temperature and humidity. The temperature outside the building is based on the database of the

Japan Meteorological Agency. The data in cleanliness-controlled areas are shown as the average values of each measurement point.

https://doi.org/10.1371/journal.pone.0236600.g002

PLOS ONE The environmental risk assessment of cell-processing facilities

PLOS ONE | https://doi.org/10.1371/journal.pone.0236600 August 5, 2020 5 / 14

it was not measured (Fig 4B). The number of colonies was well below the acceptable limit,

which was under <1 for Grade A, 10 for Grade B and 100 for Grade C and D. The entrance

times into Grade B did not correlate with the number of airborne microorganisms (S2 Fig).

PCR products from the colonies were sequenced directly, and seven bacteria, including four

gram-negative and two gram-positive bacteria, along with two fungi, were identified (S2

Table). The gross appearance of the colonies varied in color, morphology, and size (Fig 4C).

Positive colony detection rates were significantly higher at 55% or higher humidity with an

odds ratio of 2.47 (Fig 4D).

Monitoring of harmful insects

The total number of harmful insects per trap in one year shows that the majority were trapped

in the corridor outside the CPF (Fig 5A). The total number of harmful insects per trap peaked

in the corridor outside the CPF in April and August and in the laboratory outside the CPF in

August (Fig 5B). The total number of insects per trap in the corridor outside the CPF was sig-

nificantly increased when temperature outside the building was higher than 15˚C (S3 Fig).

Even in controlled areas, a small number of insects was constantly trapped between April and

December (Fig 5B). Of the total number of insects trapped, the representative six species

accounted for 92% (Fig 5C and 5D). The species of trapped insects varied by area, and the

Fig 3. Monthly transitions in the number of particles per room and the relationship between this and the number of entries to the cell-processing room. (a)

Monthly changes in the number of particles (� 0.5μm) in the room classified by each cleanliness level. The data in cleanliness-controlled areas were shown as the maximum value in each room. The dotted line indicates the acceptable limit for each room classified at each cleanliness level. (b) Relationship between integrated

number of entries to the cell-processing room (Grade B) and integrated number of particles in the cell-processing room (Grade B). p = 0.03 by Pearson’s correlation

coefficient (R = 0.68). NA: not applicable.

https://doi.org/10.1371/journal.pone.0236600.g003

PLOS ONE The environmental risk assessment of cell-processing facilities

PLOS ONE | https://doi.org/10.1371/journal.pone.0236600 August 5, 2020 6 / 14

peaks of their numbers varied by the seasons (Fig 5E and 5F). Psocoptera accounted for 41.1% and peaked in August, and only Socoptera was trapped in Grade B. Acari accounted for 40.9% and peaked in April, and they were trapped during summer in all the areas except for Grade B.

Araneae, Collembola, Psychodidae, and Blattidae accounted for 2–3% each. The others con- sisted of various species, such as Hermann (Fig 5D).

Cell production and sterility

In the one year we assessed, six cell products for transplantation were cultured in this CPF.

Their sterilities were guaranteed by the sterility test (S3 Table).

Discussion

Although cell products cannot be sterilized, their sterility must be ensured for patient safety.

To achieve the sterility of cell products, elimination of the risk of extrinsic contamination dur-

ing the process is necessary. The assessment of environmental risk factors inside and outside

CPFs will be effective to eliminate the contamination. However, these risk factors fluctuate

depending on the season, CPF location, and human intervention. In the present study, we

Fig 4. Monitoring of microorganisms and identification of species. (a) Monthly transition of colony number of airborne bacteria and fungi. Settle plate and glove print

tests were also performed by using a soybean-casein digest agar in the Grade A area. The data in cleanliness-controlled areas were shown as the maximum value in each

room. The dotted line indicates the allowable reference value in each room classified by each cleanliness level. NA: not applicable. (b) Monthly transition in falling bacteria

and bacteria attached to gloves. (c) Representative colonies of detected airborne bacteria and fungus in soybean-casein digest agar. (d) Relationship between humidity and

colony detection rate. The number of tests and the number of tests in which positive colonies were detected are shown for humidity of 55% or more and humidity less than

55%. p = 0.038 by the chi-square test.

https://doi.org/10.1371/journal.pone.0236600.g004

PLOS ONE The environmental risk assessment of cell-processing facilities

PLOS ONE | https://doi.org/10.1371/journal.pone.0236600 August 5, 2020 7 / 14

assessed environmental factors in a CPF and its adjacent GE throughout the year. Their effects

on the sterility of cell products were also assessed. Our data suggested that various environ-

mental factors could be a risk to the safety of cell products, and some of the factors interacted

with each other, such as the relationship between humidity and colony number of airborne

bacteria/fungi.

The cleanliness-controlled area in our CPF was maintained at a constant temperature, but

the humidity fluctuated by the season. The equipment for humidity control is expensive and

usually not set up in academic institutions. However, high humidity is usually the most influ-

ential factor for the growth of bacteria, fungi, and harmful insects [14–16]. In the present

study, positive colony detection rates for bacteria and fungus from airborne microorganisms

were significantly higher at humidity of 55% or higher. The fungi identified in this study, such

as Irpex lacteus, are known to grow faster under humid conditions [14], and their proliferation could affect other factors, such as the number of Psocoptera that feed on fungi [15, 17]. There- fore, high humidity could be a major threat to the safety of cell products, and the control of

humidity is an important issue for CPF management. It may thus be a solution to introduce a

simple, temporal humidity control device in humid seasons.

Fig 5. Monitoring of harmful insects. (a) Total number of insects in one year per trap. (b) Monthly transition of insects trapped outside or inside the CPF. The number

of insects is shown as maxima per trap for each month. Left y-axis indicates the number of insects per trap in the corridor outside the CPF. Right y-axis indicates the

number of insects per trap in the laboratory outside the CPF. (c) Breakdown of trapped insects. One dot represents 1%. (d) Representative macroscopic images of the

trapped insects. (e) Breakdown of trapped insects by room. (f) Heat map for monthly transitions of each representative insect per trap in each room. GE: general

environment. Cor: corridor and workspace outside the CPF. Labo: laboratory outside the CPF. Front: in front of the cell-processing facility outside the CPF. Grade D:

entrance. Grade C: first gowning room, buffer room, supply room, and de-gowning room. Grade B: second gowning room and cell-processing room. ND: not detected.

NA: not applicable.

https://doi.org/10.1371/journal.pone.0236600.g005

PLOS ONE The environmental risk assessment of cell-processing facilities

PLOS ONE | https://doi.org/10.1371/journal.pone.0236600 August 5, 2020 8 / 14

In this study, the number of particles (� 0.5μm) was within the acceptable range in all cleanliness-controlled areas throughout the year. Depending on the number of entrances

made, the integrated number of particles in Grade B increased; however, the number of entries

to the cell-processing room (Grade B) did not directly increase the number of airborne micro-

organisms. The particles in clean rooms have been reported to be dust from clothes or human

skin flakes [18]. In addition, most microorganisms attach to dusts, skin flakes, and water drop-

lets [18]. Ten to 1,000 suspended aerosol particles are considered to contain one microorgan-

ism [19, 20]. One report demonstrated that the number of particles correlated with that of

airborne microorganisms in an operating theater and an ICU [21, 22]. On the other hand,

some studies have reported that the number of particles do not correlate with the number of

microorganisms [23–25]. Our results support the latter finding. Nevertheless, during the pro-

cessing of clinical cell products, it will be essential to minimize the number of entries.

Environmental microorganisms were found in the humid season in Grade D, Grade C, and

Grade B but were not detected in Grade A. The detection of environmental microorganisms is

important to risk management in CPFs. For example, the generation of airborne microorgan-

isms in controlled areas could occur due to structural problems in CPFs associated with out-

door inflow and human problems caused by carrying in materials. Therefore, the need for the

identification of microorganisms is described in pharmacopoeias such as Japanese Pharmaco-

poeia, European Pharmacopoeia, and United States Pharmacopoeia. The microorganisms iso-

lated in this study were reported to live in the GE [26–28]. However, there were multiple

candidate species from BLAST, and they had still not been identified as one species. Our

results showed that the gross appearance of the colonies varied in color, morphology, and size,

which might be useful for the estimation of species to decide countermeasures against them.

When the same colonies are detected, it will be easy to determine the cause and appropriate

countermeasure. On the other hand, when the colonies are newly detected, the proper coun-

termeasures, such as identification of species, consideration of possible causes, and selection of

the optimal disinfection methods, are needed. The characteristics of each species (bacterium

or fungi, resistance to disinfectants, high or low growth rate, etc.) is useful for CPF manage-

ment policy.

We monitored microorganisms and identified seven bacteria, including four gram-negative

bacteria and two gram-positive bacteria. Sphingomonas paucimobilis, which lives widely in the environment, is a highly adaptable bacteria that was once recognized in a spacecraft meal [29,

30]. Ralstonia pickettii [31] and Burkholderia lata [32] have been reported to live in chlorhexi- dine disinfectant and are harmful environmental bacteria. Moraxella osloensis[33] is known as an upper respiratory tract–resident bacteria. This bacterium is a major environmental micro-

organisms, producing 4-methyl-3-hexenoic acid (4M3H), which is a cause of malodor. Methy- lobacterium suomiense was recognized in small colonies in our study. A single detection of Methylobacterium suomiense is unlikely to be an environmental risk because of their low growth rates. However, they are resistant to chlorine in tap water and surfactants, resulting in

the continuous and widespread detection of them even after treatment with usual disinfection

alone. To disinfect them, hot water at 65˚C or higher is necessary [34]. Staphylococcus hominis [35], one of the gram-positive bacteria, is also reported to be important as a nosocomial patho-

gen. Micrococcus yunnanensis [36] was isolated as environmental bacteria.

We also identified two fungi in the CPF: Cladosporium cladosporioides, which has been reported as a cause of sick building syndrome [37], and Cladosporium cladosporioides, which attaches to air conditioners [38] and present in the environment for a long time. If isolated

multiple times in CPFs, the decontamination of the entire CPF will be required. Irpex lacteus [39] is a white rot fungus distributed in the environment, such as in dead broadleaf trees.

PLOS ONE The environmental risk assessment of cell-processing facilities

PLOS ONE | https://doi.org/10.1371/journal.pone.0236600 August 5, 2020 9 / 14

Knowing these characteristics of detected microorganisms makes it possible to infer their

entry routes into CPFs and to plan disinfection strategies.

Isolated species in the present study were known to have various origins and to inhabit, not

only the environment, but also human skin [10, 40]. In fact, microorganisms derived from

human skin were detected in stem cell banks in Spain [41]. In addition, Sandle et al. reported

6,729 isolated colonies of the bacteria in the Grade A and B clean areas [42]. In their review

paper, the colonies of Micrococcus and Staphylococcus, skin resident bacteria, were detected most commonly, accounting for 50% of total colonies. The second-most common genera were

species of Bacillus, an environmental bacterium, accounting for 13%. However, other reports have demonstrated types of detected microorganisms to be rich in variety [43–45]. The differ-

ent microbiome may result from the difference in climate and processing products. In a

humid environment, such as in Singapore, various microorganisms tend to be detected [46].

Resistant bacteria have also been found in antibiotic-producing facilities [11, 47]. To maintain

CPFs properly, facility-specific evaluations based on local climatic characteristic and the pro-

cessing products are very important.

In the corridor outside the CPF, many harmful insects were trapped in spring, summer,

and autumn. Though on a smaller scale, similar results were found within the CPF. Such sea-

sonal variability is useful for predicting insect number within the CPF. Insect intrusions into

clean areas are known to be caused by humans or by decrepit facilities [48]. Since our CPF has

been completed for fewer than five years, it is highly probable that insect invasion was accom-

panied by the movement of humans and materials. Psocoptera, commonly known as booklice [49], adhere to cardboard and clothes. Psocoptera were not trapped in Grade D or Grade C but in Grade B; it is thus highly likely that they were brought in with construction materials. Insect

intrusion into a CPF may contaminate cell products because many harmful microorganisms

attach to them [50–52]. Therefore, it will be important to periodically assess the number and

characteristics of intruding insects, which can be affected by the number of entrance times, the

age of the CPF, and appropriate feedback on risk management policy.

Though bacteria, fungi, and insects were detected in Grade B–D, no contamination was

detected in the cell products during this study. One reason may be that bacteria, fungi, or

insects were not detected in Grade A. Even if contamination occurs in the cell products,

microorganisms can be detected by a sterility test of the cell products. Although this study

focused on environmental factors, management of the entire process, including the detection

of contamination and assurance of quality, is also important to realize the safety of cell

products.

We faced three limitations in this study. First, larger particles (� 10 or 20 μm) were not evaluated because our CPF was completed only five years ago and we thought that there was

no deterioration of the building. Measurement of large particles is considered useful for the

detection of building deterioration [21, 22]. Second, the difference of the distribution of micro-

organisms by season and room was not fully analyzed. If microorganisms are brought in from

the outside by clothes, improving the buffer room can be a countermeasure. If they are derived

from human skin, changing clean room clothing or modifying entrance procedures may be

useful. For them, more detailed analysis of the distribution of microorganism species in multi-

ple rooms will be needed. Third, the classification of room cleanliness did not follow ISO stan-

dards or the Japanese Pharmacopoeia. In Japan, regenerative medicine using cell products is

regulated by two laws: the Pharmaceuticals, Medical Devices and Other Therapeutic Products

Act and the Act on the Safety of Regenerative Medicine (Safety Act), which are applied in the

field of medical care and clinical research, respectively [6, 7]. Our clinical study was performed

using Safety Act standards, not ISO or Pharmacopoeia standards.

PLOS ONE The environmental risk assessment of cell-processing facilities

PLOS ONE | https://doi.org/10.1371/journal.pone.0236600 August 5, 2020 10 / 14

In conclusion, this study assessed many environmental risks inside and outside a CPF

throughout the year. Environmental factors varied with the seasons. The temperature in the

CPF was constant, but the humidity fluctuated greatly depending on the season. The number

of particles correlated with the number of entrance times. Various microorganisms were iden-

tified in the CPF, and high humidity increased their occurrence ratio. The types of trapped

insects varied by CPF area. More insects were trapped in the season when outside temperature

was higher than 15˚C. Although many risks were found in the environment around the CPF,

the sterility of the aseptic operation area and processed cells was maintained. These results pro-

vide useful data for taking the appropriate steps to keep entire CPFs clean.

Supporting information

S1 Fig. Relationship between integrated number of entries to the cell-processing room

(Grade B) and integrated number of particles (� 5μm) in the cell-processing room (Grade B). p = 0.048 by Pearson’s correlation coefficient (R = 0.64). NA: not assigned.

(PDF)

S2 Fig. The number of airborne microorganisms and entrance times to the cell-processing

room (Grade B).

(PDF)

S3 Fig. The effect of temperature outside the building on the total number of insects per

trap in the corridor outside the CPF. Data are shown by median with interquartile range. p

values were calculated by the Mann Whitney U-test.

(PDF)

S1 Table. Environment definition.

(PDF)

S2 Table. Sequence data of identified bacterium.

(PDF)

S3 Table. Sterility in cell processing products.

(PDF)

Acknowledgments

We would like to thank Ms. Jun Kusano and Ms. Akiko Hoshikawa for the management of the

Center for Cell Therapy, Ms. Mika Watanabe and Ms. Kimiko Takanashi for the management

of our laboratory, and Joe Lewis for English editing.

Author Contributions

Conceptualization: Mitsuru Mizuno.

Investigation: Mitsuru Mizuno, Ayako Tsuji, Naomi Kojima, Ken Watanabe, Norio Shimizu.

Project administration: Mitsuru Mizuno.

Supervision: Hisako Katano, Tomohiro Morio, Ichiro Sekiya.

Validation: Ichiro Sekiya.

Visualization: Mitsuru Mizuno.

Writing – original draft: Mitsuru Mizuno.

PLOS ONE The environmental risk assessment of cell-processing facilities

PLOS ONE | https://doi.org/10.1371/journal.pone.0236600 August 5, 2020 11 / 14

Writing – review & editing: Mitsuru Mizuno, Kentaro Endo, Ichiro Sekiya.

References 1. Pittenger MF, Mackay AM, Beck SC, Jaiswal RK, Douglas R, Mosca JD, et al. Multilineage potential of

adult human mesenchymal stem cells. Science. 1999; 284(5411):143–7. Epub 1999/04/02. https://doi.

org/10.1126/science.284.5411.143 PMID: 10102814.

2. Takahashi K, Tanabe K, Ohnuki M, Narita M, Ichisaka T, Tomoda K, et al. Induction of Pluripotent Stem

Cells from Adult Human Fibroblasts by Defined Factors. Cell. 2007; 131(5):861–72. Epub 2007/11/24.

https://doi.org/10.1016/j.cell.2007.11.019 PMID: 18035408.

3. Mandai M, Watanabe A, Kurimoto Y, Hirami Y, Morinaga C, Daimon T, et al. Autologous Induced Stem-

Cell–Derived Retinal Cells for Macular Degeneration. New England Journal of Medicine. 2017; 376

(11):1038–46. Epub 2017/03/16. https://doi.org/10.1056/NEJMoa1608368 PMID: 28296613.

4. Sekiya I, Koga H, Otabe K, Nakagawa Y, Katano H, Ozeki N, et al. Additional Use of Synovial Mesen-

chymal Stem Cell Transplantation Following Surgical Repair of a Complex Degenerative Tear of the

Medial Meniscus of the Knee: A Case Report. Cell Transplantation. 2019; 28(11):1445–54. Epub 2019/

07/18. https://doi.org/10.1177/0963689719863793 PMID: 31313604; PubMed Central PMCID:

PMC6802148.

5. Sekiya I, Muneta T, Horie M, Koga H. Arthroscopic Transplantation of Synovial Stem Cells Improves

Clinical Outcomes in Knees With Cartilage Defects. Clin Orthop Relat Res. 2015; 473(7):2316–26.

Epub 2015/05/01. https://doi.org/10.1007/s11999-015-4324-8 PMID: 25925939; PubMed Central

PMCID: PMC4457765.

6. Hara A, Sato D, Sahara Y. New Governmental Regulatory System for Stem Cell–Based Therapies in

Japan. Therapeutic Innovation & Regulatory Science. 2014; 48(6):681–8. Epub 2014/11/01. https://doi.

org/10.1177/2168479014526877 PMID: 30227468.

7. Konomi K, Tobita M, Kimura K, Sato D. New Japanese Initiatives on Stem Cell Therapies. Cell Stem

Cell. 2015; 16(4):350–2. Epub 2015/04/07. https://doi.org/10.1016/j.stem.2015.03.012 PMID:

25842974.

8. Mizutani M, Samejima H, Terunuma H, Kino-oka M. Experience of contamination during autologous cell

manufacturing in cell processing facility under the Japanese Medical Practitioners Act and the Medical

Care Act. Regenerative Therapy. 2016; 5:25–30. Epub 2016/07/16. https://doi.org/10.1016/j.reth.2016.

06.004 PMID: 31245497; PubMed Central PMCID: PMC6581811.

9. Ogawa Y, Mizutani M, Okamoto R, Kitajima H, Ezoe S, Kino-oka M. Understanding the formation and

behaviors of droplets toward consideration of changeover during cell manufacturing. Regenerative

Therapy. 2019. https://doi.org/10.1016/j.reth.2019.04.002 PMID: 31890765

10. Prussin AJ, Marr LC. Sources of airborne microorganisms in the built environment. Microbiome. 2015; 3

(1):78. Epub 2015/12/24. https://doi.org/10.1186/s40168-015-0144-z PMID: 26694197; PubMed Cen-

tral PMCID: PMC4688924.

11. Hamdy AM, El-massry M, Kashef MT, Amin MA, Aziz RK. Toward the Drug Factory Microbiome: Micro-

bial Community Variations in Antibiotic-Producing Clean Rooms. OMICS: A Journal of Integrative Biol-

ogy. 2018; 22(2):133–44. Epub 2017/09/06. https://doi.org/10.1089/omi.2017.0091 PMID: 28873001.

12. Takai K, Horikoshi K. Rapid Detection and Quantification of Members of the Archaeal Community by

Quantitative PCR Using Fluorogenic Probes. Applied and Environmental Microbiology. 2000; 66

(11):5066–72. Epub 2000/10/31. https://doi.org/10.1128/aem.66.11.5066-5072.2000 PMID: 11055964;

PubMed Central PMCID: PMC92420.

13. Johnson M, Zaretskaya I, Raytselis Y, Merezhuk Y, McGinnis S, Madden TL. NCBI BLAST: a better

web interface. Nucleic Acids Research. 2008; 36(Web Server):W5–W9. Epub 2008/04/29. https://doi.

org/10.1093/nar/gkn201 PMID: 18440982; PubMed Central PMCID: PMC2447716.

14. Leštan D, Lamar RT. Influence of humidity on production of pelleted fungal inoculum. World Journal of

Microbiology and Biotechnology. 1999; 15(3):349–57. https://doi.org/10.1023/a:1008983521015

15. Opit GP, Throne JE. Population Growth and Development of the Psocid Liposcelis brunnea (Psocop-

tera: Liposcelididae) at Constant Temperatures and Relative Humidities. Journal of Economic Entomol-

ogy. 2009; 102(3):1360–8. Epub 2009/07/21. https://doi.org/10.1603/029.102.0364 PMID: 19610458.

16. Schroeder H, Klotzbach H, Püschel K. Insects’ colonization of human corpses in warm and cold season.

Legal Medicine. 2003; 5:S372–S4. Epub 2003/08/26. https://doi.org/10.1016/s1344-6223(02)00135-9

PMID: 12935636.

17. Gautam SG, Opit GP, Giles KL. Population Growth and Development of the Psocid Liposcelis rufa

(Psocoptera: Liposcelididae) at Constant Temperatures and Relative Humidities. Journal of Economic

Entomology. 2010; 103(5):1920–8. Epub 2010/11/11. https://doi.org/10.1603/ec10127 PMID:

21061997.

PLOS ONE The environmental risk assessment of cell-processing facilities

PLOS ONE | https://doi.org/10.1371/journal.pone.0236600 August 5, 2020 12 / 14

18. Sandle T. Pharmaceutical Microbiology: Woodhead Publishing; 2015.

19. Gunar OV. Aspects of investigating microflora contamination of drugs (A Review). Pharmaceutical

Chemistry Journal. 2011; 45(2):93–102. https://doi.org/10.1007/s11094-011-0568-0

WOS:000293470700007.

20. Whyte W. Settling and impaction of particles into containers in manufacturing pharmacies. Journal of

Parenteral Science and Technology. 1981; 35(5):255–61. Epub 1981/09/01. PMID: 7299604.

21. Mirhoseini SH, Nikaeen M, Khanahmd H, Hatamzadeh M, Hassanzadeh A. Monitoring of airborne bac-

teria and aerosols in different wards of hospitals–Particle counting usefulness in investigation of air-

borne bacteria. Annals of Agricultural and Environmental Medicine. 2015; 22(4):670–3. Epub 2015/12/

29. https://doi.org/10.5604/12321966.1185772 PMID: 26706974.

22. Stocks GW, Self SD, Thompson B, Adame XA, O’Connor DP. Predicting bacterial populations based

on airborne particulates: A study performed in nonlaminar flow operating rooms during joint arthroplasty

surgery. American Journal of Infection Control. 2010; 38(3):199–204. Epub 2009/11/17. https://doi.org/

10.1016/j.ajic.2009.07.006 PMID: 19913327.

23. Landrin A, Bissery A, Kac G. Monitoring air sampling in operating theatres: can particle counting replace

microbiological sampling? Journal of Hospital Infection. 2005; 61(1):27–9. Epub 2005/07/13. https://doi.

org/10.1016/j.jhin.2005.03.002 PMID: 16009457.

24. Scaltriti S, Cencetti S, Rovesti S, Marchesi I, Bargellini A, Borella P. Risk factors for particulate and

microbial contamination of air in operating theatres. Journal of Hospital Infection. 2007; 66(4):320–6.

Epub 2007/07/28. https://doi.org/10.1016/j.jhin.2007.05.019 PMID: 17655973.

25. Tse H, Cristina ML, Spagnolo AM, Sartini M, Panatto D, Gasparini R, et al. Can Particulate Air Sampling

Predict Microbial Load in Operating Theatres for Arthroplasty? PLoS ONE. 2012; 7(12):e52809. Epub

2013/01/04. https://doi.org/10.1371/journal.pone.0052809 PMID: 23285189; PubMed Central PMCID:

PMC3528722.

26. Gomez-Silvan C, Leung MHY, Grue KA, Kaur R, Tong X, Lee PKH, et al. A comparison of methods

used to unveil the genetic and metabolic pool in the built environment. Microbiome. 2018; 6(1):71. Epub

2018/04/18. https://doi.org/10.1186/s40168-018-0453-0 PMID: 29661230; PubMed Central PMCID:

PMC5902888.

27. Kelley ST, Gilbert JA. Studying the microbiology of the indoor environment. Genome Biology. 2013; 14

(2):202. Epub 2013/03/22. https://doi.org/10.1186/gb-2013-14-2-202 PMID: 23514020; PubMed Cen-

tral PMCID: PMC3663111.

28. Merino N, Zhang S, Tomita M, Suzuki H. Comparative genomics of Bacteria commonly identified in the

built environment. BMC Genomics. 2019; 20(1):92. Epub 2019/01/30. https://doi.org/10.1186/s12864-

018-5389-z PMID: 30691394; PubMed Central PMCID: PMC6350394.

29. Ryan MP, Adley CC. Sphingomonas paucimobilis: a persistent Gram-negative nosocomial infectious

organism. Journal of Hospital Infection. 2010; 75(3):153–7. Epub 2010/05/04. https://doi.org/10.1016/j.

jhin.2010.03.007 PMID: 20434794.

30. Koenig DW, and Pierson D. L. Microbiology of the space shuttle water system. Water Science and

Technology. 1997; 35(11–12):59–64. Epub 1997/01/01. https://doi.org/10.1016/s0273-1223(97)00235-

7 PMID: 11540417.

31. Ryan M, Pembroke J, Adley C. Ralstonia pickettii: a persistent Gram-negative nosocomial infectious

organism. Journal of Hospital Infection. 2006; 62(3):278–84. Epub 2005/12/13. https://doi.org/10.1016/

j.jhin.2005.08.015 PMID: 16337309.

32. Leong LEX, Lagana D, Carter GP, Wang Q, Smith K, Stinear TP, et al. Burkholderia lata Infections from

Intrinsically Contaminated Chlorhexidine Mouthwash, Australia, 2016. Emerging Infectious Diseases.

2018; 24(11):2109–11. Epub 2018/10/20. https://doi.org/10.3201/eid2411.171929 PMID: 30334703;

PubMed Central PMCID: PMC6199994.

33. Kubota H, Mitani A, Niwano Y, Takeuchi K, Tanaka A, Yamaguchi N, et al. Moraxella Species Are Pri-

marily Responsible for Generating Malodor in Laundry. Applied and Environmental Microbiology. 2012;

78(9):3317–24. Epub 2012/03/01. https://doi.org/10.1128/AEM.07816-11 PMID: 22367080; PubMed

Central PMCID: PMC3346475.

34. Yano T, Kubota H, Hanai J, Hitomi J, Tokuda H. Stress Tolerance of Methylobacterium Biofilms in Bath-

rooms. Microbes and Environments. 2012; 28(1):87–95. Epub 2012/12/05. https://doi.org/10.1264/

jsme2.me12146 PMID: 23207727; PubMed Central PMCID: PMC4070686.

35. Chaves F, Garcia-Alvarez M, Sanz F, Alba C, Otero JR. Nosocomial Spread of a Staphylococcus homi-

nis subsp. novobiosepticus Strain Causing Sepsis in a Neonatal Intensive Care Unit. Journal of Clinical

Microbiology. 2005; 43(9):4877–9. Epub 2005/09/08. https://doi.org/10.1128/JCM.43.9.4877-4879.

2005 PMID: 16145165; PubMed Central PMCID: PMC1234135.

PLOS ONE The environmental risk assessment of cell-processing facilities

PLOS ONE | https://doi.org/10.1371/journal.pone.0236600 August 5, 2020 13 / 14

36. Zhao GZ, Li J, Qin S, Zhang YQ, Zhu WY, Jiang CL, et al. Micrococcus yunnanensis sp. nov., a novel

actinobacterium isolated from surface-sterilized Polyspora axillaris roots. International Journal of Sys-

tematic and Evolutionary Microbiology. 2009; 59(10):2383–7. https://doi.org/10.1099/ijs.0.010256–0

37. Cooley JD, Wong WC, Jumper CA, Straus DC. Correlation between the prevalence of certain fungi and

sick building syndrome. Occupational and Environmental Medicine. 1998; 55(9):579–84. Epub 1998/

12/23. https://doi.org/10.1136/oem.55.9.579 PubMed Central PMCID: PMC1757636. PMID: 9861178

38. Ahearn DG, Simmons RB, Switzer KF, Ajello L, Pierson DL. Colonization byCladosporium spp. of

painted metal surfaces associated with heating and air conditioning systems. Journal of Industrial

Microbiology. 1991; 8(4):277–80. https://doi.org/10.1007/BF01576067 PMID: 11536513

39. Novotný Č, Erbanová P, Cajthaml T, Rothschild N, Dosoretz C, ŠaŠek V. Irpex lacteus, a white rot fun- gus applicable to water and soil bioremediation. Applied Microbiology and Biotechnology. 2000; 54

(6):850–3. Epub 2001/01/11. https://doi.org/10.1007/s002530000432 PMID: 11152080.

40. Grice EA, Segre JA. The skin microbiome. Nature Reviews Microbiology. 2011; 9(4):244–53. Epub

2011/03/17. https://doi.org/10.1038/nrmicro2537 PMID: 21407241; PubMed Central PMCID:

PMC3535073.

41. Cobo F, Concha Á. Environmental microbial contamination in a stem cell bank. Letters in Applied Micro-

biology. 2007; 44(4):379–86. Epub 2007/04/03. https://doi.org/10.1111/j.1472-765X.2006.02095.x

PMID: 17397475

42. Sandle T. A Review of Cleanroom Microflora: Types, Trends, and Patterns. PDA Journal of Pharmaceu-

tical Science and Technology. 2011; 65(4):392–403. Epub 2012/02/02. https://doi.org/10.5731/pdajpst.

2011.00765 PMID: 22293526.

43. Pacchioni F, Esposito A, Giacobazzi E, Bettua C, Struffi P, Jousson O. Air and waterborne microbiome

of a pharmaceutical plant provide insights on spatiotemporal variations and community resilience after

disturbance. BMC Microbiology. 2018; 18(1):124. Epub 2018/10/05. https://doi.org/10.1186/s12866-

018-1267-8 PMID: 30285625; PubMed Central PMCID: PMC6171249.

44. Park HK, Han JH, Joung Y, Cho SH, Kim SA, Kim SB. Bacterial diversity in the indoor air of pharmaceu-

tical environment. Journal of Applied Microbiology. 2014; 116(3):718–27. Epub 2013/12/10. https://doi.

org/10.1111/jam.12416 PMID: 24314234.

45. Vijayakumar R, Sandle T. A review on biocide reduced susceptibility due to plasmid-borne antiseptic-

resistant genes—special notes on pharmaceutical environmental isolates. Journal of Applied Microbiol-

ogy. 2018; 126(4):1011–22. Epub 2018/10/03. https://doi.org/10.1111/jam.14118 PMID: 30276940.

46. Balasubramanian R, Nainar P, Rajasekar A. Airborne bacteria, fungi, and endotoxin levels in residential

microenvironments: a case study. Aerobiologia. 2011; 28(3):375–90. https://doi.org/10.1007/s10453-

011-9242-y

47. Mora M, Mahnert A, Koskinen K, Pausan MR, Oberauner-Wappis L, Krause R, et al. Microorganisms in

Confined Habitats: Microbial Monitoring and Control of Intensive Care Units, Operating Rooms, Clean-

rooms and the International Space Station. Frontiers in Microbiology. 2016; 7:1573. Epub 2016/10/30.

https://doi.org/10.3389/fmicb.2016.01573 PMID: 27790191; PubMed Central PMCID: PMC5061736.

48. Damani N. Manual of Infection Prevention and Control2019.

49. Berry C, Meyer JM, Hoy MA, Heppner JB, Tinzaara W, Gold CS, et al. Bark-Lice, Book-Lice or Psocids

(Psocoptera). Encyclopedia of Entomology2008. p. 381–99.

50. Gurung K, Wertheim B, Falcao Salles J. The microbiome of pest insects: it is not just bacteria. Entomo-

logia Experimentalis et Applicata. 2019; 167(3):156–70. https://doi.org/10.1111/eea.12768

51. Nadarasah G, Stavrinides J. Insects as alternative hosts for phytopathogenic bacteria. FEMS Microbiol-

ogy Reviews. 2011; 35(3):555–75. Epub 2011/01/22. https://doi.org/10.1111/j.1574-6976.2011.00264.

x PMID: 21251027.

52. Ruiu L. Insect Pathogenic Bacteria in Integrated Pest Management. Insects. 2015; 6(2):352–67. Epub

2015/10/16. https://doi.org/10.3390/insects6020352 PMID: 26463190; PubMed Central PMCID:

PMC4553484.

PLOS ONE The environmental risk assessment of cell-processing facilities

PLOS ONE | https://doi.org/10.1371/journal.pone.0236600 August 5, 2020 14 / 14

Copyright of PLoS ONE is the property of Public Library of Science and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.