Revised Estimates for the Number of Human and Bacteria Cells in the Body — TEI-tagged view

Ron SenderpersName, Shai FuchspersName, Ron MilopersName (2016date). “Revised Estimates for the Number of Human and Bacteria Cells in the Body.” PLoS Biologytitle·venue.

Object of study — the number of human and bacterial cellsterm·def in the body, and the provenance of the values the literature repeats for them

Research question — what the two counts actually are when the calculation is documented and its uncertainty derived, and what the widely-cited 10:1 ratio between them rests on

Contribution — revised estimates with the calculation logic, sources and uncertainty rangesterm fully documented, replacing the 10:1 ratio with about 1:1 - and the observation that the difference between those two numbers is a definition

Full text — tagged

The reference list is not part of the text we hold for this paper. Our copy was extracted from the publisher's page and the bibliography did not survive that extraction, so the article below ends before its references. Nothing has been removed by us.

Revised Estimates for the Number of Human and Bacteriaterm·def Cellsterm·def in the Body. PLoS Biol 14(8): e1002533. doi:10.1371/journal.pbio.1002533 Published: August 19, 2016 Copyright: © 2016 SenderpersName 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.

Funding: This work was funded by the European Research Council (Project NOVCARBFIX 646827, https://erc.europa.eu/funding-and-grants); Dana and Yossie Hollander; Helmsley Charitable Foundation; The Larson Charitable Foundation; The Estate of David Arthur Barton; The Anthony Stalbow Charitable Trust, and Stella Gelerman, Canada. RM is the Charles and Louise Gartner professional chair and an EMBO young investigator program member. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing Interests: The authors have declared that no competing interests exist.

Abbreviations: CV, coefficient of variation; FISH, fluorescent in situ hybridization; SD, standard deviation; SA, surface area; SEM, Standard Error of the Mean.

Reported values in the literature on the number of cellsterm·def in the body differ by orders of magnitude and are very seldom supported by any measurements or calculations. Here, we integrate the most up-to-date information on the number of human and bacterial cellsterm·def in the body. We estimate the total number of bacteriaterm·def in the 70 kg "reference manterm" to be 3.8(cid:1)1013. For human cellsterm·def, we identify the dominant role of the hematopoietic lineageterm to the total count ((cid:3)90%) and revise past estimates to 3.0(cid:1)1013 human cellsterm·def. Our analysis also updates the widely-cited 10:1 ratio, showing that the number of bacteriaterm·def in the body is actually of the same order as the number of human cellsterm·def, and their total mass is about 0.2 kg.

Introduction How many cellsterm·def are there in the human body? Beyond order of magnitude statements that give no primary reference or uncertainty estimates, very few detailed estimates have been performed (the one exception [1] is discussed below). Similarly, the ubiquitous statements regarding 1014–1015 bacteriaterm·def residing in our body trace back to an old back-of-the-envelopeterm·mth calculation [2–4].

The aim of this study is to critically revisit former estimates for the number of human and bacterial cellsterm·def in the human body. We give up-to-date detailed estimates where the calculation logic and sources are fully documented and uncertainty rangesterm are derived. By updating the cellterm·def counts in the body, we also revisit the 10:1 value that has been so thoroughly repeated as to achieve the status of an established common knowledge fact [4]. This ratio was criticized recently in a letter to the journal Microbe [5], but an alternative detailed estimate that will give concrete values and estimate the uncertainty rangeterm·stat is needed. Here, we provide an account of the methodologies employed hitherto for cellterm·def count and revise past estimates. Doing so, we repeat and reflect on the assumptions in previous back-of-the-envelopeterm·mth calculations, also known as Fermi problemsterm·mth. We find such estimates as effective sanity checks and a way to improve our quantitative understanding in biology.

A major part of the available literature used in the derivation of human cell numbersterm·var was based on cohorts of exclusively or mostly men, and as we use these sources, our analysis starts PLOS Biology | DOI:10.1371/journal.pbio.1002533 August 19, 2016 1 / 14 with adult men. As discussed below, relatively modest quantitative differences apply for women due to changes in characteristic body mass, blood volume, and the genital microbiota. For our analysis, we used the definition of the standard reference manterm as given in the literature [6] as: "Reference Manterm is defined as being between 20–30 years of age, weighing 70 kg, is 170 cm in height.” Our analysis revisits the estimates for the number of microbial cellsterm·def, human cellsterm·def, and their ratio in the body of such a standard man.

We begin our analysis by revisiting the number of bacteriaterm·def through surveying earlier sources, comparing counts in different body organs and finally focusing on the content of the colonterm. We then estimate the total number of human cellsterm·def in the body, comparing calculations using a "representative" cellterm·def size to aggregation by cellterm·def type. We then contrast the cell numberterm·var distribution by tissue type to the mass distribution. In closing, we revisit the ratio of bacterial to human cellsterm·def and evaluate the effect of gender, age, and obesity.

Results Origin of Prevalent Claims in the Literature on the Number of Bacterial Cellsterm·def in Humans Microbes are found throughout the human body, mainly on the external and internal surfaces, including the gastrointestinal tract, skin, saliva, oral mucosa, and conjunctiva. Bacteriaterm·def overwhelmingly outnumber eukaryotes and archaea in the human microbiome by 2–3 orders of magnitude [7,8]. We therefore sometimes operationally refer to the microbial cellsterm·def in the human body as bacteriaterm·def. The diversity in locations where microbes reside in the body makes estimating their overall number daunting. Yet, once their quantitative distribution shows the dominance of the colonterm as discussed below, the problem becomes much simpler. The vast majority of the bacteriaterm·def reside in the colonterm, with previous estimates of about 1014 bacteriaterm·def [2], followed by the skin, which is estimated to harbor ~1012 bacteriaterm·def [9].

As we showed recently [4], all papers regarding the number of bacteriaterm·def in the human gastrointestinal tract that gave reference to the value stated could be traced to a single back-of-theenvelope estimate [3]. That order of magnitude estimate was made by assuming 1011 bacteriaterm·def per gram of gut content and multiplying it by 1 liter (or about 1 kg) of alimentary tractterm capacity. To get a revised estimate for the overall number of bacteriaterm·def in the human body, we first discuss the quantitative distribution of bacteriaterm·def in the human body. After showing the dominance of gut bacteriaterm·def, we revisit estimates of the total number of bacteriaterm·def in the human body.

Distribution of Bacteriaterm·def in Different Human Organs Table 1 shows typical order of magnitude estimates for the number of bacteriaterm·def that reside in different organs in the human body. The estimates are based on multiplying measured concentrations of bacteriaterm·def by the volume of each organ [9,10]. Values are rounded up to give an order of magnitude upper boundterm.

Although the bacterial concentrations in the saliva and dental plaque are high, because of their small volume the overall numbers of bacteriaterm·def in the mouth are less than 1% of the colonterm bacteriaterm·def number. The concentration of bacteriaterm·def in the stomachterm and the upper 2/3 of the small intestine (duodenum and jejunum) is only 103–104 bacteriaterm·def/mL, owing to the relatively low pH of the stomachterm and the fast flow of the content through the stomachterm and the small intestine [10]. Table 1 reveals that the bacterial content of the colonterm exceeds all other organs by at least two orders of magnitude. Importantly, within the alimentary tractterm, the colonterm is the only substantial contributor to the total bacterial population, while the stomachterm and small intestine make negligible contributions.

PLOS Biology | DOI:10.1371/journal.pbio.1002533 August 19, 2016 2 / 14 Table 1. Bounds for bacteriaterm·def number in different organs, derived from bacterial concentrations and volume.

Location Colonterm (large intestine) Dental plaque Ileum (lower small intestine) Saliva Skin Stomachterm Duodenum and Jejunum (upper small intestine) Typical concentration of bacteriaterm·def (1) (number/mL content) 1011 1011 108 109 <1011 per m2 (3) 103–104 103–104 Volume (mL) 400 (2) <10 400 (5) <100 1.8 m2 (4) 250 (5)–900 (6) 400 (5) Order of magnitude bound for bacteriaterm·def number 1014 1012 1011 1011 1011 107 107 (1) Except for skin, concentrations are according to [9]. For the skin, we used bacterial areal density and total skin surface to reach an upper bound. (2) See derivation in section below. (3) Skin surface bacteriaterm·def density is taken from [11]. (4) Skin area calculated as inferred from standard formula by DuBois for the body surface area [12]. (5) Volume of the organs of the gastrointestinal tract is derived from weights taken from [13] by assuming content density of 1.04 g/mL [6]. (6) Higher value is given in [14].

doi:10.1371/journal.pbio.1002533.t001 Revisiting the Original Back-of-the-Envelopeterm·mth Estimate for the Number of Bacteriaterm·def in the Colonterm The primary source for the often cited value of ~1014 bacteriaterm·def in the body dates back to the 1970s [3] and only consists of a sentence-long “derivation,” which assumes the volume of the alimentary tractterm to be 1 liter, and multiplies this volume by the number density of bacteriaterm·def, known to be about 1011 bacteriaterm·def per gram of wet content. Such estimates are often very illuminating, yet it is useful to revisit them as more empirical data accumulates. This pioneering estimate of 1014 bacteriaterm·def in the intestine is based on assuming a constant bacterial density over the 1 liter of alimentary tractterm volume (converting from volume to mass via a density of 1 g/mL). Yet, the parts of the alimentary tractterm proximal to the colonterm contain a negligible number of bacteriaterm·def in comparison to the colonterm content, as can be appreciated from Table 1. We thus conclude that the relevant volume for the high bacteriaterm·def density of 1011 bacteriaterm·def/g is only that of the colonterm. As discussed in Box 1, we integrated data sources on the volume of the colonterm to arrive at 0.4 L.

The Total Number of Bacteriaterm·def in the Body We are now able to repeat the original calculation for the number of bacteriaterm·def in the colonterm [3]. Given 0.9(cid:1)1011 bacteriaterm·def/g wet stool as derived in Box 2 and 0.4 L of colonterm, we find 3.8(cid:1)1013 bacteriaterm·def in the colonterm with a standard error uncertainty of 25% and a variation of 52% SD over a population of 70 kg males. Considering that the contribution to the total number of bacteriaterm·def from other organs is at most 1012, we use 3.8(cid:1)1013 as our estimate for the number of bacteriaterm·def across the whole body of the "reference manterm." We note that the uncertainty estimate value takes into account known variation in the colonterm volume, bacteriaterm·def density, etc., but cannot account for unquantified systematic biases. One prominent such bias is the knowledge gap on differences between the actual bacteriaterm·def density in the colonterm, with all its spatial heterogeneity, and the measurements of concentration in feces, which serve as the proxy for estimating bacteriaterm·def number.

What is the total mass of bacteriaterm·def in the body? From the total colonterm content of about 0.4 kg and a bacteriaterm·def mass fraction of about one-half [21,24], we get a contribution of about 0.2 kg PLOS Biology | DOI:10.1371/journal.pbio.1002533 August 19, 2016 3 / 14 Box 1. The Volume of the Human Colonterm Content This is a critical parameter in our calculation. We used a value of 0.4 L based on the following studies (see also S1 Data, tab ColonContent). The volume of the colonterm content of the reference adult man was previously estimated as 340 mL (355 g at density of 1.04 g/ mL [6]), based on various indirect methods including flow measurements, barium meal X-ray measurements and postmortem examination [13]. A recent study [15] gives more detailed data about the volume of undisturbed colonterm that was gathered by MRI scans. The authors report a height-standardized colonic inner volume for males of 97 ± 24 mL/ m3 (where the best fit was found when dividing the colonic volume by the cube of the height). Taking a height of 1.70 m for the reference manterm [6], we arrive at a colonterm volume of 480 ± 120 mL (where unless noted otherwise ± refers to the standard deviation [SD]). This volume includes an unreported volume of gas and did not include the rectum. Most recently, studies analyzing MRI images of the colonterm provided the most detailed and complete data. The inner colonterm volume in that cohort was 760 mL in total [16,17]. This cohort was, however, significantly taller than the reference manterm. Normalizing for height, we arrive at 600 mL total volume for a standard man. In order to deduct the volume occupied by gas, stool fraction in this report was estimated at (cid:3)70% of colonterm volume leading to 430 mL of standardized wet colonterm content. Therefore, this most reliable analysis together with earlier studies support an average value of about 0.4 L.

We can sanity-check this volume estimate by looking at the volume of stool that flows through the colonterm. An adult human is reported to produce on average 100–200 grams of wet stool per day [18]. The colonic transit time is negatively correlated with the daily fecal output, and its normal values are about 25–40 hours [18,19]. By multiplying the daily output and the colonterm transit time, we thus get a volume estimate of 150–250 mL, which is somewhat lower than but consistent with the values above, given the uncertainties and very crude estimate that did not account for water in the colonterm that is absorbed before defecation. To summarize, the volume of colonterm content as evaluated by recent analyses of MRI images is in keeping with previous estimates and fecal transit dynamics. Values for a reference adult man averaged 0.4 L (standard error of the mean [SEM] 17%, coefficient of variation [CV] 25%), which will be used in calculations below. Following a typical meal, the volume changes by about 10% [15], while each defecation event reduces the content by a quarter to a third [18].

(wet weight) from bacteriaterm·def to the overall mass of the colonterm content. Given the dominance of bacteriaterm·def in the colonterm over all other microbiota populations in the body, we conclude that there is about 0.2 kg of bacteriaterm·def in the body overall. Given the water content of bacteriaterm·def, the total dry weight of bacteriaterm·def in the body is about 50–100g. This value is consistent with a parallel alternative estimate for the total mass of bacteriaterm·def that multiplies the average mass of a gut bacterium of about 5 pg (wet weight, corresponding to a dry weight of 1–2 pg, see S1 Appendix) with the updated total number of bacteriaterm·def. We note that this empirically observed average gut bacterium is several times bigger than the conveniently chosen “standard” 1 μm3 volume and 1 pg wet mass bacterium often referred to in textbooks. The total bacteriaterm·def mass we find represents about 0.3% of the overall body weight, significantly updating previous statements that 1%–3% of the body mass is composed of bacteriaterm·def or that a normal human hosts 1–3 kg of bacteriaterm·def [25].

PLOS Biology | DOI:10.1371/journal.pbio.1002533 August 19, 2016 4 / 14 Box 2. Concentration of Bacteriaterm·def in the Colonterm The most widely used approach for measuring the bacterial cellterm·def density in the colonterm is by examining bacteriaterm·def content in stool samples. This assumes that stool samples give adequate representation of colonterm content. We return to this assumption in the discussion. The first such experiments date back to the 1960s and 1970s [20,21]. In those early studies, counting was based on direct microscopic clump counts from diluted stool samples. Later experiments [22,23] used DAPI nucleic acid staining and fluorescent in situ hybridization [FISH] to bacterial 16S RNA. Values are usually reported as bacteriaterm·def per gram of dry stool. For our calculation, we are interested in the bacteriaterm·def content for the wet rather than dry content of the colonterm. To move from bacteriaterm·def/g dry stool to bacteriaterm·def/g wet stool we use the fraction of dry matter as reported in each article. Table 2 reports the values we extracted from 14 studies in the literature and translated them to a common basis enabling comparison.

The Number of Human Cellsterm·def in a “Standard” Adult Male Many literature sources make general statements on the number of cellsterm·def in the human body ranging between 1012 to 1014 cellsterm·def [26,27]. An order of magnitude back-of-the-envelopeterm·mth argument behind such values is shown in Box 3.

Table 2. Values of bacteriaterm·def density in stool as reported in several past articles.

Article Author Houte & Gibbons Moore & Holdeman Holdeman, Good & Moore Stephen & Cummings Langendijk et al.

Franks et al.

Simmering & Kleessen Tannock et al.

Harmsen, Raangs, He, Degener & Welling Zoetendal et al Zhong et al.

Thiel & Blaut He et al.

Uyeno, Sekiguchi & Kamagata Mean Year 1966 1974 1976 1980 1995 1998 1999 2000 2002 2002 2004 2005 2008 2008 bac. #/g dry stool (x1011) dry matter as % of stool bac. #/g wet stool (x1011) CV(%) 4.1 2.9 4.8 2.1 2.9 1.5 3.5 1.5 22% 31% 29%(1) - - 30% 23% 25% - 27% ± 2% 3.2 1.1 1.3 1.2 2.7 0.74(2) 1.3(2) 0.95 0.62 0.77(2) 0.35 0.87 0.39(2) 0.44 0.92 ± 19% 53% 78% 66% 25% 26% 39% 44% 40% 38% 24% 73% 53% 43% 34% 46% Full references are provided in Table A in S1 Appendix. Mean bacteriaterm·def number is calculated using the geometric mean to give robustness towards outlier values. Values quoted directly from the articles are written in bold, values derived by us are written in italic. Values reported with more than two significant digits are rounded to two significant digits as the uncertainty makes such overspecification nonsensible. ± standard error of the mean. (1) Value for [21] derived from their Table 1. (2) From derivation, assuming the averaged dry matter fraction of 27%. From the measurements collected in Table 2, we calculated the representative bacteriaterm·def concentration in the colonterm by two methods, yielding very close values: the geometric mean is 0.92(cid:1)1011 (SEM 19%) bacteriaterm·def per gram of wet stool, while the median of the values is 0.91(cid:1)1011 (SEM 19% by bootstrapping, see methods in S1 Appendix). The variation across the population, given by the average CV, is 46%.

doi:10.1371/journal.pbio.1002533.t002 PLOS Biology | DOI:10.1371/journal.pbio.1002533 August 19, 2016 5 / 14 Box 3. Order of Magnitude, Naïve Estimate for the Number of Human Cellsterm·def in the Body Assume a 102 kg man, consisting of “representative” mammalian cellsterm·def. Each mammalian cellterm·def, using a cellterm·def volume of 1,000–10,000 μm3 , and a cellterm·def density similar to that of water, −11 kg. We thus arrive at 1013–1014 human cellsterm·def in total in the body, will weigh 10 as shown in Fig 1. For these kind of estimates, where cellterm·def mass is estimated to within an order of magnitude, factors contributing to less than 2-fold difference are neglected. Thus, we use 100 kg as the mass of a reference manterm instead of 70 kg and similarly ignore the contribution of extracellular mass to the total mass. These simplifications are useful in making the estimate concise and transparent.

−12–10 An alternative method that does not require considering a representative "average" cellterm·def systematically counts cellsterm·def by type. Such an approach was taken in a recent detailed analysis [1]. The number of human cellsterm·def in the body of each different category (by either cellterm·def type or organ system) was estimated. For each category, the cellterm·def count was obtained from a literature reference or by a calculation based on direct counts in histological cross sections. Summing over a total of 56 cellterm·def categories [1] resulted in an overall estimate of 3.7(cid:1)1013 human cellsterm·def in the body (SD 0.8(cid:1)1013, i.e., CV of 22%).

Updated Inventory of Human Cellsterm·def in the Body In our effort to revisit the measurements cited, we employed an approach that tries to combine the detailed, census approach with the benefits of a heuristic calculation used as a sanity check. We focused on the six cellterm·def types that were recently identified [1] to comprise 97% of the human cellterm·def count: red blood cellsterm (accounting for 70%), glial cellsterm·def (8%), endothelial cellsterm·def (7%), dermal fibroblasts (5%), plateletsterm (4%), and bone marrow cellsterm·def (2%). The other 50 cellterm·def types account for the remaining 3%. In four cases (red blood cellsterm, glial cellsterm·def, endothelial cellsterm·def, and dermal fibroblasts), we arrived at revised calculations as detailed in Box 4.

Our revised calculations of the number of glial cellsterm·def, endothelial cellsterm·def, and dermal fibroblast yield only 0.9(cid:1)1012 cellsterm·def, in contrast to 7.5(cid:1)1012 cellsterm·def in the previous estimate. This leaves us Fig 1. Back of the envelopeterm·mth estimate of the number of cellsterm·def in an adult human body based on a characteristic volume and mass.

doi:10.1371/journal.pbio.1002533.g001 PLOS Biology | DOI:10.1371/journal.pbio.1002533 August 19, 2016 6 / 14 Box 4. Revised Estimates for the Number of Red Blood Cellsterm, Glial Cellsterm·def, Endothelial Cellsterm·def, and Dermal Fibroblasts The largest contributor to the overall number of human cellsterm·def are red blood cellsterm. Calculation of the number of red blood cellsterm was made by taking an average blood volume of 4.9 L (SEM 1.6%, CV 9%) multiplied by a mean RBC count of 5.0(cid:1)1012 cellsterm·def/L (SEM 1.2%, CV 7%) (see Table 3 and S1 Data). The latter could be verified by looking at your routine complete blood count, normal values range from 4.6–6.1(cid:1)1012 cellsterm·def/L for males and 4.2– 5.4(cid:1)1012 cellsterm·def/L for females. This led to a total of 2.5(cid:1)1013 red blood cellsterm (SEM 2%, CV 12%). This is similar to the earlier report of 2.6(cid:1)1013 cellsterm·def [1].

The number of glial cellsterm·def was previously reported as 3(cid:1)1012 [1]. This estimate is based on a 10:1 ratio between glial cellsterm·def and neurons in the brain. This ratio of glia:neurons was held as a broadly accepted convention across the literature. However, a recent analysis [28] revisits this value and, after analyzing the variation across brain regions, concludes that the ratio is close to 1:1. The study concludes that there are 8.5(cid:1)1010 glial cellsterm·def (CV 11%) in the brain and a similar number of neurons and so we use these updated values here.

The number of endothelial cellsterm·def in the body was earlier estimated at 2.5(cid:1)1012 cellsterm·def (CV 40%), based on the mean surface area of one endothelial cellterm·def [1] and the total surface area of the blood vessels, based on a total capillary length of 8(cid:1)109 cm. We could not find a primary source for the total length of the capillary bed and thus decided to revisit this estimate. We used data regarding the percentage of the blood volume in each type of blood vessels [29]. Using mean diameters for different blood vessels [30], we were able to derive (S1 Data) the total length of each type of vessel (arteries, veins, capillaries, etc.) and its corresponding surface area. Dividing by the mean surface area of one endothelial cellterm·def [31], we derive a reduced total estimate of 6(cid:1)1011 cellsterm·def.

The number of dermal fibroblasts was previously estimated to be 1.85(cid:1)1012 [1], based on multiplying the total surface area (SA) of the human body (1.85 m2 [32]) by the areal density of dermal fibroblasts [33]. We wished to incorporate the dermal thickness (d) into the calculation. Dermal thickness was directly measured at 17 locations throughout the body [34], with the mean of these measurements yielding 0.11±0.04 cm. The dermis is composed of two main layers: papillary dermis (about 10% of the dermis thickness) and reticular dermis (the other 90%) [35]. The fibroblast density is greater in the papilpap. of 106 cellsterm·def/cm2 (with 100 μm thickness lary dermis, with a reported areal density, σ of papillary, giving 108 cellsterm·def/cm3) [33]. The fibroblast density in the middle of the dermis was reported to be about 3(cid:1)106 cellsterm·def/cm3 [36], giving an areal density of σ ret. = 3(cid:1)105 cellsterm·def/ ret.) = 1.85(cid:1)104 cm2 (106 + 3(cid:1)105) cm2. Combining these we find: Nder.fib. = SA(cid:1)(σ cellsterm·def/cm2 = 2.6(cid:1)1010 cellsterm·def. After this 100-fold decrease in number, dermal fibroblasts are estimated to account for only (cid:3)0.05% of the human cellterm·def count.

pap. + σ with 3.0(cid:1)1013 human cellsterm·def in the 70 kg “reference manterm,” with a calculated 2% uncertainty and 14% CV. We note that the uncertainty and CV estimates might be too optimistically low, as they are dominated by the reported high confidence of studies dealing with red blood cellsterm but may underestimate systematic errors, omissions of some cellterm·def types, and similar factors that are hard to quantify. In Fig 2, we summarize the revised results for the contribution of the different cellterm·def types to the total number of human cellsterm·def. Categories contributing >0.4% in cellterm·def count are PLOS Biology | DOI:10.1371/journal.pbio.1002533 August 19, 2016 7 / 14 Fig 2. The distribution of the number of human cellsterm·def by cellterm·def type. Representation as a Voronoi tree map where polygon area is proportional to the number of cellsterm·def. Visualization performed using the online tool at http://bionic-vis.biologie.unigreifswald.de/.

doi:10.1371/journal.pbio.1002533.g002 presented. All the other categories sum up to about 2% together. We find that the body includes only 3(cid:1)1012 non-blood human cellsterm·def, merely 10% of the total updated human cellterm·def count. The visualization in Fig 2 highlights that almost 90% of the cellsterm·def are estimated to be enucleated cellsterm·def (26(cid:1)1012 cellsterm·def), mostly red blood cellsterm and plateletsterm, while the other (cid:3)10% consist of (cid:3)3(cid:1)1012 nucleated cellsterm·def. The striking dominance of the hematopoietic lineageterm in the cellterm·def count (90% of the total) is counterintuitive given the composition of the body by mass. This is the subject of the following analysis.

Mass-Centered Approach as Sanity Check for Cellterm·def Count It is prudent in making such estimates to approach the analysis from different angles. In that spirit, we now ask does the cumulative mass of the cellsterm·def counted fall within the expected range for a reference adult? To properly tackle that question, we first need to state what the anticipated result is, i.e., total body cellterm·def mass. For a reference manterm mass of 70 kg, 25% is extracellular fluid [37], another 7% is extracellular solids [37], thus we need to account for (cid:3)46 kg of cellterm·def mass (including fat).

A comprehensive systematic source for the composition of total cell massterm (rather than total cellterm·def count) is the Report of the Task Group on Reference Manterm [6], which gives values for the mass of the main tissues of the human body. This mass per tissue analysis includes both intra- and extracellular components. To distinguish between intra- and extracellular portions of each tissue, we can leverage total body potassium measurements [38] as detailed in S1 Appendix. Fig 3 compares the main tissues that contribute to the human body, in terms of cell numberterm·var and masses.

A striking outcome of this juxtaposition is the evident discordance between contributors to total cell massterm and to cell numberterm·var. The cellterm·def count is dominated by red blood cellsterm (84%), among PLOS Biology | DOI:10.1371/journal.pbio.1002533 August 19, 2016 8 / 14 Fig 3. Distribution of cell numberterm·var and mass for different cellterm·def types in the human body (for a 70 kg adult man). The upper bar displays the number of cellsterm·def, while the lower bar displays the contribution from each of the main cellterm·def types comprising the overall cellular body mass (not including extracellular mass that adds another (cid:3)24 kg). For comparison, the contribution of bacteriaterm·def is shown on the right, amounting to only 0.2 kg, which is about 0.3% of the body weight.

doi:10.1371/journal.pbio.1002533.g003 the smallest cellterm·def types in the human body with a volume of about 100 μm3. In contrast, 75% of total cell massterm is composed of two cellterm·def types, fat cellsterm (adipocytesterm) and muscle cellsterm (myocytesterm), both large cellsterm·def (usually >10,000 μm3 by volume) that constitute only a minute fraction ((cid:3)0.2%) of total cell numberterm·var. At the other extreme, bacteriaterm·def have a minor contribution in terms of mass, but a cellterm·def count comparable to all human cellsterm·def combined, as discussed above. The mass balance accounts well for all expected body mass, giving support to our analysis. The option of overlooking a collection of very small cellsterm·def numerous enough to alter the total cellterm·def count is further discussed in the S1 Appendix.

The Ratio of Bacteriaterm·def to Human Cellsterm·def in the Adult Body With the revised estimates for the number of human (3.0(cid:1)1013) and bacterial cellsterm·def (3.8(cid:1)1013) in the body (the numerator and denominator of the B/H ratio), we can give an updated estimate of B/ H = 1.3, with an uncertainty of 25% and a variation of 53% over the population of standard 70 kg males. This B/H valueterm of about 1:1 (with the associated uncertainty rangeterm·stat) should replace the 10:1 or 100:1 values that are stated in the literature until more accurate measurements become available.

We note that if one chooses to compare the number of bacteriaterm·def in the human body (3.8(cid:1)1013) to the number of nucleated human cellsterm·def ((cid:3)0.3(cid:1)1013), the ratio will be about 10:1. This is because the dominant population of non-nucleated red blood cellsterm is not included in the calculation. We note that this ratio is the result of both the number of bacteriaterm·def and the number of nucleated human cellsterm·def in the body being several times lower than in the original 1970s estimate (that did not restrict the analysis to nucleated cellsterm·def). The issue of whether cellsterm·def without a nucleus should be included or discarded in the calculation of the number of human cellsterm·def, and thus the B/ H ratio, seems to be a question of definition. We view red blood cellsterm as bona fide cellsterm·def, as their name suggests. But it is also plausible to choose not to include them as some may think of them as “bags full of hemoglobin.” Inclusion of plateletsterm in the count, which corresponds to their inclusion in previous counts, is also potentially disputable but has only a minor quantitative effect. Indeed, this opens an interesting tangential discussion on what should be defined as a cellterm·def.

Variations in the Ratio of Bacteriaterm·def to Human Cellsterm·def across Population Segments After reviewing the B/H ratio for the “reference manterm,” we now generalize our results by addressing other segments of the population. Looking at our estimate, we identify four main parameters that dominate the calculation: PLOS Biology | DOI:10.1371/journal.pbio.1002533 August 19, 2016 9 / 14 Table 3. B/H ratio for different population. See Table B in S1 Appendix for full