www.nature.

com/scientificreports

Corrected: Publisher Correction

OPEN Signal transmission through

elements of the cytoskeleton form
an optimized information network
Received: 5 October 2018
Accepted: 25 March 2019 in eukaryotic cells
Published online: 16 April 2019
B. R. Frieden1 & R. A. Gatenby2

Multiple prior empirical and theoretical studies have demonstrated wire-like flow of electrons and
ions along elements of the cytoskeleton but this has never been linked to a biological function. Here
we propose that eukaryotes use this mode of signal transmission to convey spatial and temporal
environmental information from the cell membrane to the nucleus. The cell membrane, as the interface
between intra- and extra-cellular environments, is the site at which much external information is
received. Prior studies have demonstrated that transmembrane ion gradients permit information
acquisition when an environmental signal interacts with specialized protein gates in membrane ion
channels and producing specific ions to flow into or out of the cell along concentration gradients. The
resulting localized change in cytoplasmic ion concentrations and charge density can alter location and
enzymatic function of peripheral membrane proteins. This allows the cell to process the information
and rapidly deploy a local response. Here we investigate transmission of information received and
processed in and around the cell membrane by elements of the cytoskeleton to the nucleus to alter gene
expression. We demonstrate signal transmission by ion flow along the cytoskeleton is highly optimized.
In particular, microtubules, with diameters of about 30 nm, carry coarse-grained Shannon information
to the centrosome adjacent to the nucleus with minimum loss of input source information. And,
microfilaments, with diameters of about 4 nm, transmit maximum Fisher (fine-grained) information
to protein complexes in the nuclear membrane. These previously unrecognized information dynamics
allow continuous integration of spatial and temporal environmental signals with inherited information
in the genome.

Survival and proliferation of living systems require them to continuously acquire, process, and respond to infor-
mation1 from the environment for threats, opportunities, or (in the case of multicellular organisms) instruc-
tions from local tissue2. The cell membrane, as the interface between a cell and its environment, is the site at
which much of this environmental information is received. Some environmental changes, such as perturbations
in osmolarity, temperature and pH, typically affect all regions of the cell membrane equally and simultaneously.
Other information, such as (1) the presence of a potential predator or food source; or, (2) in the tissue of a highly
ordered multicellular organism, the space-time location of a target cell must be spatially and temporally resolved.
A prior study3 demonstrated that the steep transmembrane ion gradients in eukaryotes are critical for receiv-
ing and processing environmental information. Information is received when some perturbation causes the pro-
tein gates in transmembrane ion channels to open. The subsequent flow of one or more ions into or out of the cell
along these pre-existing electro-chemical gradients can induce local changes (Fig. 1) that promote an adaptive
(both fast and targeted) cellular response. For example, an outflow of K+ (the dominant mobile cation in the cyto-
plasm) may reduce the shielding of fixed negative charges on the inner leaf of the cell membrane, enhancing elec-
trostatic forces for attracting or repelling charges on nearby macromolecules. Furthermore, the activity of many
enzymes is dependent on cation concentrations, so that a local fluctuation may increase or decrease their activity4.
The role of transmembrane ion movement has been recognized for decades as the mechanism of nerve con-
duction5. As expressed by the Hodgkin-Huxley (H-H) equations6, propagating depolarization waves along an

1
College of Optical Science, University of Arizona, Tucson, AZ, USA. 2Department of Integrated Mathematical
Biology, Moffitt Cancer Center, Tampa, FL, USA. Correspondence and requests for materials should be addressed to
R.A.G. (email: [email protected])

Scientific Reports | (2019) 9:6110 | https://doi.org/10.1038/s41598-019-42343-2 1

Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/ www.nature.com/scientificreports

Figure 1. Information dynamics in and around the cell membrane and transmission to central organelles.
The resting state of the membrane, with large transmembrane concentration gradients of K+, Na+ and
Cl− is shown in the upper left panel. In the lower left panel, an environmental signal causes the gates in
transmembrane K+ channels to open. This allows rapid flow of K+ out of the cell briefly altering the ion
concentrations and charge balance in the cytoplasm. Prior studies have shown these ion dynamics alter
localization and function of peripheral membrane proteins permitting analysis of and response to the
environmental perturbation. This ion flux in the cytoplasm adjacent to the cell membrane can also enter the
channel of adjacent microtubules which allows transmission of coarse-grained information to the centrosome
(see Fig. 3). The ion flux can also change the electrical potential at the distal end of a microfilament allowing
electron flow along the wire-like structure transmitting (Fig. 3) fine-grained information to a protein complex
in the nuclear membrane which can alter gene expression and chromosomal location.

Figure 2. Distribution of microfilaments and microtubules in eukaryotes. Immunohistochemistry stains

showing the distribution of microtubules (green) and microfilaments (red) within normal fibroblasts.

axon are generated by sequential transmembrane flows of ions through membrane channels. There is one H-H
equation for each ion. Our model proposes3 that the ion dynamics that produce a traveling depolarization wave
in neurons are, in fact, a specialized application of membrane information dynamics that are universally obeyed
by eukaryotes.
Here we address the question of how environmental information that is transmitted through the cell mem-
brane through ion fluxes is communicated internally to other components of the cell. We expect that many

Scientific Reports | (2019) 9:6110 | https://doi.org/10.1038/s41598-019-42343-2 2

Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/ www.nature.com/scientificreports

Figure 3. Cytoskeletal structures as information conduits. An environmental perturbation that causes

transmembrane flow of K+ out of the adjacent cytoplasm (Fig. 2) generates a transient ion gradient along the
length of the hollow core of a microtubule or a potential gradient along the microfilament which forms a wire-
like conductor for ion flow.

environmental perturbations (e.g. a localized mechanical deformation by a small environmental object) may
only elicit and require a local response. However, some signals received at the membrane, because of their con-
tent, amplitude, or spatiotemporal frequency, may require a global (or ‘coordinated’) cellular response including
increased energy production in the mitochondria7,8 and changes in gene expression9 or translation within the
nucleus and endoplasmic reticulum10.
We propose that information encoded in local fluxes of ions in the cytoplasm adjacent to the cell membrane
can be transmitted to other organelles by elements of the intracellular cytoskeleton, its microtubules and micro-
filaments (Figs 2 and 3). These form organized networks throughout eukaryotic cells that are often complex and
dynamic11 (Fig. 2).
Although the elements of the cytoskeleton are primarily involved in cellular shape and movement8, they can
also serve as both biomechanical and electrical conduits of information that can alter gene expression9 and chro-
mosomal location9,12. The ability of microfilaments and microtubules to conduct electrons and ions has been
extensively documented13. Microfilaments are actin polymers which maintain high levels of negative surface
charges permitting highly dynamical interactions with both within the microfilament and in exchange with cyto-
plasmic counterions14,15. A number of studies have demonstrated charge centers with corresponding counter
ion clouds along the microfilament axis permitting ionic waves propagating along its long axis15–18. This con-
ductance takes on a specific form as Cantiello et al. demonstrated actin filaments propagate electrical signal via
soliton waves18 so that the signal is virtually lossless. Ionic conduction along the length of microtubules has also
been observed and S. The precise mechanism is not clear19 but may include diffusion along the central channel20
and ion redistribution along the microtubule as a result of variations in cation (Na, K, Ca) flux through nano-
pores along the microtubule wall21. Furthermore, microtubules are capable of amplifying ionic signal waves22.
There is also experimental evidence that microtubules can regulate VDAC ion channels in the mitochondria23
and that microtubules can influence24 and be influenced by the extracellular matrix25 (ECM), so that there is a
dynamic and ongoing exchange of information between the cytoskeleton and cell exterior. Finally, recent work
by Santelices et al.26 has experimentally demonstrated that microtubule responses to AC electric signals are fre-
quency and ion concentration dependent.
The cytoskeleton fibers are often arrayed in organized patterns along the radius of the cell from the nuclear
membrane to the cell membrane (Fig. 1). Furthermore, the proximal ends of microtubules, which typically join
together in the centrosome27 adjacent to the nuclear membrane and microfilaments, are often bound to multipro-
tein structures (e.g. the KASH-SUN complex28) in the outer nuclear membrane. Molecular tethers are, thereby,
formed that have diverse functions, e.g. gene transcription, as well as transmission of forces for chromosome
movements and nuclear migration. Furthermore, actin filaments directly alter nuclear pore ion channel activity,
thus altering the ionic milieu of the nucleoplasm29.
Hence we investigate the potential of microtubules and microfilaments to act as information conductors that
link the cell membrane with central cellular structures including the nucleus, mitochondria and endoplasmic
reticulum. These thereby form a distributed information network of conductors.

Summary of the Biological Model

Our biological information model proposes environmental perturbations can be detected by specialized gates
on membrane ion channels. When the gate opens, ions specific to the channel will flow along concentration
gradients into or out of the cell. Depending on the number of open channels, and the duration of that open state,
the cytoplasm adjacent to the channels will undergo a rapid change in ion concentrations, charge density, and

Scientific Reports | (2019) 9:6110 | https://doi.org/10.1038/s41598-019-42343-2 3

Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/ www.nature.com/scientificreports

osmolarity. Once the gates are closed, rapid re-equilibration will occur through diffusion from adjacent regions of
the cytoplasm and activation of transmembrane ion pumps.
When the transmembrane ion flows within one or a few channels (measured to be about 105 ions/second per
channel) are very brief, such as depicted by the Hodgkin-Huxley (H-H) equations, we expect the consequences of
this change will be entirely localized to the region of the membrane. As described by H-H (see reference 5 below)
this ion flux will produce local changes in location and function of peripheral membrane proteins that constitute
a rapid local response to the perturbation, so that information transmission to other cellular organelles is not
necessary. This is similar to autonomic functions in multicellular organisms that deal with isolated, transient
perturbations through local reflexes.
In contrast, we expect some environmental information will cause gate openings of multiple ion channels,
and/or will maintain the channels in an open state for a longer period of time. This will increase the amplitude,
time, and spatial distribution of changes in cytoplasmic ion concentrations. As we will see, it will also maximize
the received information about the environment which, in turn, maximizes speed of signal flow and speed of
decoding. We hypothesize that optimal cell function and survival will additionally necessitate that this informa-
tion be transmitted to other organelles so as to elicit a more global cellular response. Examples are increased ATP
production by the mitochondria and alterations in gene expression and translation.
In general, optimal communication in complex networks will integrate coarse- and fine-grained dynamics.
Coarse-grained conduits transmit information from larger temporal and spatial scales to allow more rapid and
efficient processing of large scale perturbations30,31. Fine-grained information dynamics focus upon the transmis-
sion of finer spatial and temporal scales32. Here, we hypothesize that local ion changes in the cytoplasm adjacent
to open transmembrane channels can alter the terminal ends of local cytoskeletal structures. As shown in Figs 2
and 3, microtubules are relatively large11 (~30 nm in diameter) tubes with a hollow center. Empirical studies have
demonstrated that signals in the form of ionic waves can be transmitted along the course of a microtubule22. We
view information transmitted by the microtubule as “coarse grained30” because:

1. The microtubule is sufficiently large that it will primarily detect ion changes that occur within a cross sec-
tional region similar to its diameter (~30 nm).
2. Most microtubules connect to the centrosome, which is typically positioned adjacent to the nuclear mem-
brane with which it communicates33. While most recognized for its role in microtubule organization and
spindle assembly, the centrosome it is also associated with molecules associated with diverse cellular func-
tions including cell-cycle progression, checkpoint control, ubiquitin-mediated degradation33,34 and protein
kinase A (PKA)35 which has diverse regulatory functions in cell metabolism36. Thus, the centrosome, in
effect, will tend to “average” signals from multiple regions over time and communicates this summation to
the nucleus.
3. The microtubule has a number of additional electro-magnetic37,38 properties that could additionally inte-
grate the activities of multiple microtubules within the cell cytoplasm or even extend into the extracellular
matrix or adjacent cells.

In contrast, microfilaments (diameter ~4 nm) can communicate in fine detail39. Prior studies have demon-
strated that microfilaments, which are composed of actin with highly negative surface charge, are highly conduc-
tive and, in fact, have been used as nanowires that respond to osmotic and electric potential differences18,22,38,40,41.
Because of its small diameter, we propose microfilaments, in contrast to microtubules, will transmit fine-grained
information showing fluctuations on the order of microseconds (“fine-grained”). Hence we propose that changes
in cytoplasmic ion concentrations near the membrane end of the microfilament–with the other end continuing
to have the usual ion concentration distribution–will generate a potential difference across the microfilament
“wire,” resulting in ion flow. Microfilaments typically attach to multi-protein complexes in the nuclear membrane
which have been shown to control both gene expression and chromosomal locations. Thus, in this component of
the system, signal transmission is both very rapid–“tunable” (in the sense that the electron flow is dependent on
the potential along the length of the microtubule)–and well-resolved both spatially and temporally at both signal
transmission and reception.
In summary, we propose that elements of the cytoskeleton mediate biomechanical activity and can likewise
carry information. In addition, both microtubules and microfilaments connect with mitochondria and the endo-
plasmic reticulum, and can course along the cytoplasm adjacent to the cell membrane. This allows a broad dis-
tribution of information signals to enter many regions of the cell (and even flow to other cells). However, for
simplicity, we here focus on information transmission between specifically the cell membrane and the nuclear
membrane.

Principles of Information Transmission

The dynamics governing information transmission have been extensively investigated by the pioneering work of
Fisher, Shannon and others (see below). Ideal information transmission occurs when the receiver obtains pre-
cisely the information that was sent from its source. This is because, any channel carrying a signal from a sender
to a receiver cannot (by definition of a channel) convey more information than is contained in the source signal.
Instead, there is inevitable loss of source Shannon information en route during the process of encoding, transmis-
sion, reception and decoding of the message. Therefore, minimizing such loss is the realistic goal of such a system.
Note that this ignores the evolutionary cost to the system of acquiring the message. Instead, it tacitly assumes
every such possible message to be acquired with equal cost, and focuses upon the issue of how well the system
can respond to that message, regardless of cost. The assumption is that, realistically, the external system activities
producing the message are not controllable by the observer. Hence, from the point of view of survival, evolutionary
dynamics will optimize an organism’s response to any signal that can affect its fitness. This requires a maximum

Scientific Reports | (2019) 9:6110 | https://doi.org/10.1038/s41598-019-42343-2 4

Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/ www.nature.com/scientificreports

likelihood estimation42,43 approach rather than, e.g., one seeking a posterior mean, since the latter would require
knowing the probability of each such possible message.
Meeting this goal of maximum likelihood estimation of external state properties in living systems requires:
(a) minimization of the occurrence of data errors and (as below) (b) maximization of the rate of received infor-
mation. Such a scenario is very beneficial for purposes of biological survival in the random, and possibly hostile,
environment we are assuming. As will be seen, ion transmission through microtubules or microfilaments achieves
this dual aim (a), (b).
As found below, coarse grained and fine grained information acquisition are subject to two different princi-
ples of minimum information loss (i) I − J = minimum and (ii)SI − SJ = minimum, where subscripts I and J refer,
respectively, to information received and at the environmental source. In this paper we focus on the environmen-
tal information that enters the cell in the form of a transmembrane flow of one or more ions. How well can that
information be received? Also, can that information be transmitted optimally (as above) to other regions of the
cell via the cytoskeleton network?
In principle (i), I and J are, respectively, levels of temporal Fisher information when just received and just sent,
over the continuous, total time interval (0, T) of flow; likewise in principle (ii) for the Shannon informations SI
and SJ. Then, by either principle the information loss is minimized. The particular choice of principle, (i) or (ii), is
governed by the fineness of the spatial structure forming the information conduit, as follows.
In systems with true fine structure (order of 1–5 nm) information I is the ion’s level of Fisher information44,
with J the equivalent physical information45. Depending on case, J could be the sum of all ion concentrations,
mean times within the system; etc., affecting I. Or it could be the information as represented in a conjugate space
to t, such as energy-momentum in quantum-relativistic problems46–48.
Thus, principle (i), I−J = minimum, operates on the finest level of cellular structure. It has been termed46
EPI (“extreme physical information”) and applies to the finest ion signal flow, through microfilaments of actin.
Operating on this finest scale allows principle (i) to even give rise to quantum effects45, such as the Schrodinger
wave equation (where, in particular, J is the mean kinetic energy47; although this is not the case here). But in all
cases, J in principle (i) is the largest possible value of I. It results that principle (i) produces a maximized value
for the Fisher information Principle (i) is shown later to give rise to the well-known Hodgkin-Huxley equations.
By comparison, principle (ii)SI − SJ = minimum, where SI and SJ are, respectively, the levels of Shannon infor-
mation (in bits) as, respectively, received and sent. This applies when the system signals are coarse grained. It was
first applied to telephone communication (real flow of electron charges through real, metal wires, represented by
microfilaments of actin here), by C.E. Shannon49. For such coarse-grained microtubules (order of 25–100 nm)
principle (ii) becomes ΔS ≡ SI − SJ = minimum, one of minimum loss of Shannon information (ii). Principle (ii)
is, thus, a non-quantum, coarse-grained theory. It results in the highest possible delivered information from an
arbitrary source message in the environment. In fact principle (ii) directly derives as a coarse-grained version of
principle (i) (see below). As a verification, one solution to problem (ii) of coarser ion flow is found to obey the
classical Hodgkin-Huxley equations6 (see Eq. 5, below).
These calculations based on minimum information-loss principles (i) and (ii) indicate that real biological
systems, such as the neuron of the giant squid obey ion flow rates delivering optimum levels of acquired infor-
mation3. However, this ignores the issue of how the cells use information in the decoded messages to optimize
survival and fitness2. To answer it, the benefits of acquiring and communicating each component of available infor-
mation of threat or benefit in the environment would have to be weighed against the resources needed to maintain
the molecular machinery necessary to cope with it. To do such a calculation would require knowing unknowable
probabilities of unknown possible threats. This is, again, why we can only calculate the response to an arbitrary
message of threat or benefit as in the preceding section.
Ideally, to maintain critically important cell functions loss of information in transmission must be minimized
using a fine grain network. Analogously, a person translating a book from ancient hieroglyphics to English does
not have to do it perfectly (i.e. with zero error) to produce a generally useful translation for consumption by the
general public. However, translation of information regarding, for example, dates and places may be essential
for historians or archaeologists. Here, the translation must be as close to perfect as possible (i.e. with minimum
error).
Since principle (ii) of minimum loss of Shannon information derives (in Sec. 5) as a coarse-grained approxi-
mation to the EPI principle (i), and since all eukaryotes contain tubulin, probably all have likewise evolved out of
the principle of minimum loss of Shannon information. As will be seen this is a necessary condition for natural
selection.
In summary, we present a variational principle of biophysics that governs intracellular information flow based
upon the granularity of the conduit through which the signal flows. We propose that the principle of minimum
loss of Fisher information applies when fine grained information is transmitted electron flow through microfil-
aments (sometimes called ‘actin wires16’). In contrast, the principle of minimum loss of Shannon information
during transmission governs coarse grained information carried by ion flow through microtubules. Importantly,
however, we note that “minimum loss” also means, in a positive sense, maximum gain and thus can have the effect
of increasing fitness. As we noted, these metrics of information transmission are directly related: When Fisher
information undergoes coarse graining, it becomes proportional to Shannon information. This is, then, an impor-
tant bridge between the discrete and continuous aspects of living systems.

Fisher Information
All information forms used in this paper ultimately arise out of Fisher information. By definition, this
obeys44,45,47,48,50:

Scientific Reports | (2019) 9:6110 | https://doi.org/10.1038/s41598-019-42343-2 5

Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/ www.nature.com/scientificreports

2
 da 
I=4 ∫ dt  dt  , a = a(t) ≡ p(t ) ,
(1)
where p = p(t) is the probability density on position t for the ion and a(t) is defined as its (real) amplitude. All
integrals are over a fixed time interval 0 ≤ t ≤ T of observation.For now, we notice that the form of Eq. (1) is also
that of a Lagrangian L in integral ∫ dtL, and this is conventionally varied as δ∫ dtL = 0 to derive45 the quantum
mechanics obeyed by amplitude law a(t) in scenarios of fine structure 1–5 nm or, alternatively, by classical flow of
p(t) in coarser structure of size 25–100 nm. The emphasis in this paper is on the latter (classical domain)
behavior.
The information I defined by Eq. (1) is also conventionally used to measure, by the relation44,45,48
2
emin = 1/I , (2)
the minimum possible mean-squared error e of any estimate of the true time t0 based on its repeated meas-
2

urement tn = t0 + errori n = 1, … N. Thus, I has the significance of defining how well a quantity on the continuum
(here of time values t) can be known. (Notice that In Eq. (2), the larger information I is the smaller is the rms error
emin, as one would expect of an information measure). Relation (2) has been the basis for usual past uses of Fisher
information I.
By comparison, over the recent two decades another, completely different use of information I has arisen. Its
aim is, not to merely measure particular values tn of an observable phenomenon, as above, but rather to estimate
the actual probability law p(t) governing t in the unknown phenomenon (of physics47, econophysics, biology51,
cancer52 growth, chemistry, etc.45,53). The present paper extends these calculations to problems of ion- or electron
transmission using principles (i) or (ii).

Its Physical Manifestation J, EPI Principle

This is by using a principle of extreme “physical information” I − J,
I − J = minimum (3)
through variation of p(t). Although both I and J are metrics of information hey differ in meaning. From the
( )
2
factor da in Eq. (1), information I governs the amount of ‘slope’ or ‘roughness’ in both probability law p(t) and
dt
its amplitude law a(t). Also, by Eq. (2), I governs how accurately an unknown coordinate t can be known. The
other quantity J in (3) defines the meaning of the information I as a physical quantity. Their difference I−J is called
the ‘physical information,’ since it measures how much net information I−J contributes to the physical effect.
Hence Eq. (3) expresses a principle of extreme net physical information (EPI). In our cellular scenario it, in fact,
represents a scenario of minimum lost, temporal Fisher information. What does this mean?
Quantity I−J is always convex, so it defines a minimum value when varied mathematically. Such minimization
means I≈J, i.e. the theoretical information tends to equal its physical manifestation. In fact, in quantum scenar-
ios45,48 I = J, meaning the entire physical manifestation J of information I (here the energy) goes into forming the
observable information I.
Meanwhile, in our case of cellular information flow J is input as the mean time of ion flow from the cytoplasm
adjacent to the cell membrane to its other end at the nuclear membrane. This choice of J was that of Hodgkin and
Huxley6. Hence, principle (3) provides the solution for amplitudes a = a(t) in signal transmission of fine struc-
tured time scale. The principle (3) of Extreme Physical Information or EPI has been used to derive most of text-
book physics46 and some laws of biology54 including a prediction of power law growth m(t ) = m(0)t Ø for
early-growth stages of breast cancer52,55 where the constant Ø = 1.618034 that has been confirmed in multiple
studies using mammography data.

Transition to Principle of Minimum Kullback-Leibler (KL) Divergence

We demonstrate above that the Fisher information-based EPI principle (3) is directly applicable to problems of
unknown ion rate functions p(t) or a(t) on the continuum of t, typified by spatial observations on the nanometer
(fine) scale 1–10 nm (case of microfilaments). For this scale of problem the information I was found to be Fisher’s,
given by Eq. (1).
However, as noted above, information transmission via microtubules instead requires the ions to flow along
a much larger structure (~30 nm in diameter). This is a coarse-grained problem and our goal is to find the
‘coarse-grained’ rates p(t), q(t). To do this, we examine a transition from the fine-scaled principle (3) to the cor-
responding coarse-grained problem.
We start by using identity p(t) = a2(t) to express Eq. (1) it in terms of the probabilities p(t),

(dp /dt )2
I= ∫ dt p
.
(4)
However, for application of the EPI principle (3) to this discrete problem of the soma, we need the form of
the information (4) where differentials dt are regarded as small but finite, “granular” changes Δt. The latter are
defined as follows.
In principle the finest time interval dt in principle (4) is of size zero. However, in practice it is the finite time
interval Δt during which the ion is located within the cell membrane. This is time between the instant that the
ion just enters the cell membrane, from the outside environment, to the time that it just emerges from the cell

Scientific Reports | (2019) 9:6110 | https://doi.org/10.1038/s41598-019-42343-2 6

Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/ www.nature.com/scientificreports

membrane and enters the cell cytoplasm. On this basis, how large an information value I is delivered by the ion i
to the observer during this time interval? Let it obey probability qi(t) in the outside environment. Then probability
pi(t) after entering the cytoplasm a short time Δt later is pi(t + t). Then the information I in Eq. (4) is easily found
to approximately obey45:
∞
I = 2/(Δt 2) ∫0 dtpi (t )ln(pi (t )/qi (t )) ≡ HKL(pi ||qi ) = minimum, i = 1, … , N .
(5)
This is the information delivered to the observer as limited by the granular nature of the ion and the medium
(cell membrane) it passes through. Notation HKL(pi||qi) denotes the Kullback-Leibler (K-L) divergence53,56 (or
‘distance’) between probabilities pi(t) and qi(t) (these are, equivalently, ion flow rates, since the random variable
is the time t).
Factor 2/(Δt2) in (5) shows that observing the time with a finer (smaller) “grain size” Δt gives greater informa-
tion I in cell structure such as a microtubule. This is intuitively correct. More importantly, Eq. (5) also generally
represents the loss in Shannon information53 for an ion i passing through the membrane regarded as an infor-
mation channel. By (5) the loss is, then, explicitly, minimized in this coarse-grained scenario. In summary, the
information I is identically the loss in Shannon information45 during the flow from along the microtubule, and
this loss is minimized. This is central to the information-based approach here. In summary, both principles (i) and
(ii) define scenarios of minimum loss of information, although of different types – Fisher information in (i), and
Shannon information in (ii).
Finally, even if grain size Δt is not very small, by the approximate nature of principle (5) it may be used as sim-
ply a first-order (in change Δt) approximation to the general principle (3). That is, a Shannon information-based
calculation (ii) is always at least an approximate solution to the Fisher-information-based one (i).

Insertion of Prior Knowledge

The minimum in Eq. (3) is to be obtained in the presence of prior knowledge J about the trajectories. These must,
e.g., obey normalization. But the key prior knowledge was found by Hodgkin and Huxley6 to be the mean times τi,
i = 1, …, N for ions in the system (cochlea). The minimum value in Eq. (5) is also constrained by this knowledge.
Using these as additive Lagrange constraints J during minimization of I, principle (5) becomes one of constrained
KL (Kullback-Leibler) divergence53
∞
I−J= ∫0 dtpi (t) ln(pi (t)/qi (t)) + λ1 ∫ dt [pi (t ) − 1]
+λ 2 ∫ dt[qi (t ) − 1] + λ 3∫ dttpi (t ) − τi  = min
  (6)
The first right-hand term is the KL divergence in (5). The terms in λ1 and λ2 express normalization of prob-
ability densities pi(t) and qi(t) or, more generally, the total presence of ion i over time interval (0, T). Notice that
Eq. (6) is of the same general form as the EPI principle (3) and, hence, is called the KLmin principle. It has been
found that this KLmin principle derives the equations governing the actual ion trajectories pi(t), i = 1, …, N, i.e. the
Hodgkin-Huxley equations6. Further significance is that, because of the required minimum value in principle (5)
the Shannon information is maximized for the transit of each ion i (i.e. its information loss is minimized).
Although, as was noted, this theory holds for transmembrane ion flows carrying a propagating signal along
neurons, it holds as well for ion flux within microtubules.

Discussion
Transmission of information from the cell membrane to the nucleus and other central organelles following ligand
binding to membrane receptors has been extensively studied. Typically, the receptor binding triggers (often
through intermediate proteins) phosphorylation of messenger proteins in the cytoplasm that then travel to other
organelles. These information pathways are widely investigated and play important roles in cellular function as
well as cancer development. However, we note that signal transmission via 3 dimensional random walk (notably,
not obeying principle (5) of maximum acquired information level) inevitably suffers significant degradation of
information regarding the time and location (on the cell membrane) of the perturbation57.
Under many circumstances, this lossy information is, nevertheless, sufficient to elicit a necessary response58,59
and has the benefit of low energy consumption60. However, acquiring even more information about the location
and time of perturbation may be essential under some circumstances such as locating a predator or a potential
food source by single cell eukaryotes, or moving to a correct cellular location within the highly ordered 3 dimen-
sional structure of tissue in a multicellular organism. In prior work, we have proposed that eukaryotes use the
difference in ion concentrations in the extra-cellular and intra-cellular fluid as a mechanism to receive, process,
and respond to a wide range of perturbations in the environment. Indeed, the value of maintaining this mem-
brane information receiver is evident in studies that show about 40% of the energy budget in eukaryotic cells61 is
consumed by the ATP-dependent membrane ion pumps that maintain the gradient.
We have proposed cellular information dynamics include ion-specific transmembrane channels which permit
communication in the form of ion flows between the environment and the cell. This occurs when the specialized
gate (there are well over a hundred different types of gates) is induced to open by the environmental perturbation.
Thus, the ion flow in toto actually represents an optimized response (5) to the nature of the perturbation, as well as
its time and place. The subsequent local processing and response are described above.
In experimental studies of a highly specialized application of this principle – the ion flow carrying a traveling
wave in neurons – it is clear that the change in local ion concentrations is rapidly dissipated through diffusion
from adjacent regions of the cytoplasm and activation of the transmembrane ion pumps. However, in normal cell

Scientific Reports | (2019) 9:6110 | https://doi.org/10.1038/s41598-019-42343-2 7

Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/ www.nature.com/scientificreports

function, we anticipate that this information, although transient and spatially localized, will in many situations
need to be communicated to other components of the cell.
Here we focus on the potential role of microfilaments and microtubules in this communication network.
These long linear polymers are often arrayed in organized patterns and frequently observed to be oriented along
the radius of the cell from the nuclear membrane to the cell membrane. The potential for both microtubules and
microfilaments to transmit signals via ion conduction has been extensively investigated both theoretically and
experimentally. Here we propose that extracellular information that is received by specialized gates in membrane
ion channels and transmitted through ion transmembrane ion fluxes can be propagated by microfilaments and
microtubules to the nucleus and other internal organelles such as the mitochondria. The specific properties of
microfilaments and microtubules allow them to carry fine-grain or coarse-grain information respectively. In the
case of microtubules, which typically converge on the centrosome, this coarse grain information allows rapid
assessment of the overall state of the environment over time. Or in the case of microfilaments, which typically link
to protein complexes on the nuclear membrane, the fine-grain information can convey, to the nucleus, detailed
information about the spatial and temporal variations of the environment. In prior work3,66,67 we found these
information dynamics to be highly optimized.
We note that such optimization, obeying information maximization principles (i) or (ii), give rise to optimally
fast and effective responses to environmental challenges and benefits. Hence, these are a necessary condition for
natural selection to have taken place (and continue).
Finally, we note that our analysis ignores possible communication between individual microfilaments and
microtubules. In reality, microfilaments and microtubules frequently interface through direct physical contact
and cross-linking proteins. It is also likely that the elements of the cytoskeleton can interact in complex way with
molecular transduction pathways. This suggests a complex network for signal transmission and analysis that
permits rich information dynamics that likely augments and modifies the more well-known and studied infor-
mation found molecular pathways (e.g. the MAPK pathway) that carry information following ligand binding to
a membrane receptor to the nucelus.

References
1. Farnsworth, K. D., Nelson, J. & Gershenson, C. Living is Information Processing: From Molecules to Global Systems. Acta Biotheor
61, 203–222, https://doi.org/10.1007/s10441-013-9179-3 (2013).
2. Gatenby, R. A. & Frieden, B. R. Information theory in living systems, methods, applications, and challenges. Bull Math Biol 69,
635–657, https://doi.org/10.1007/s11538-006-9141-5 (2007).
3. Gatenby, R. A. & Frieden, B. R. Cellular information dynamics through transmembrane flow of ions. Sci Rep 7, 15075, https://doi.
org/10.1038/s41598-017-15182-2 (2017).
4. Page, M. J. & Di Cera, E. Role of Na+ and K+ in enzyme function. Physiol Rev 86, 1049–1092, https://doi.org/10.1152/
physrev.00008.2006 (2006).
5. Hodgkin, A. L. The relation between conduction velocity and the electrical resistance outside a nerve fibre. J Physiol 94, 560–570
(1939).
6. Hodgkin, A. L. & Huxley, A. F. A quantitative description of membrane current and its application to conduction and excitation in
nerve. J Physiol 117, 500–544 (1952).
7. Boldogh, I. R. & Pon, L. A. Interactions of mitochondria with the actin cytoskeleton. Biochim Biophys Acta 1763, 450–462, https://
doi.org/10.1016/j.bbamcr.2006.02.014 (2006).
8. Foissner, I. Microfilaments and microtubules control the shape, motility, and subcellular distribution of cortical mitochondria in
characean internodal cells. Protoplasma 224, 145–157, https://doi.org/10.1007/s00709-004-0075-1 (2004).
9. Zheng, B., Han, M., Bernier, M. & Wen, J. K. Nuclear actin and actin-binding proteins in the regulation of transcription and gene
expression. FEBS J 276, 2669–2685, https://doi.org/10.1111/j.1742-4658.2009.06986.x (2009).
10. Gurel, P. S., Hatch, A. L. & Higgs, H. N. Connecting the cytoskeleton to the endoplasmic reticulum and Golgi. Curr Biol 24,
R660–R672, https://doi.org/10.1016/j.cub.2014.05.033 (2014).
11. Alberts, B. Molecular biology of the cell. (Garland Pub., 1983).
12. Starr, D. A. & Fridolfsson, H. N. Interactions between nuclei and the cytoskeleton are mediated by SUN-KASH nuclear-envelope
bridges. Annu Rev Cell Dev Biol 26, 421–444, https://doi.org/10.1146/annurev-cellbio-100109-104037 (2010).
13. Woolf, N. J., Priel, A. & Tuszynski, J. A. Nanoneuroscience: structural and functional roles of the neuronal cytoskeleton in health and
disease. (Springer, 2009).
14. Tang, J. X. & Janmey, P. A. The polyelectrolyte nature of F-actin and the mechanism of actin bundle formation. J Biol Chem 271,
8556–8563 (1996).
15. Tuszynski, J. A., Portet, S., Dixon, J. M., Luxford, C. & Cantiello, H. F. Ionic wave propagation along actin filaments. Biophys J 86,
1890–1903, https://doi.org/10.1016/S0006-3495(04)74255-1 (2004).
16. Patolsky, F., Weizmann, Y. & Willner, I. Actin-based metallic nanowires as bio-nanotransporters. Nat Mater 3, 692–695, https://doi.
org/10.1038/nmat1205 (2004).
17. Hunley, C., Uribe, D. & Marucho, M. A multi-scale approach to describe electrical impulses propagating along actin filaments in
both intracellular and in vitro conditions. RSC Advances 8, 12017–12028, doi:0.1039/C7RA12799E (2018).
18. Lin, E. C. & Cantiello, H. F. A novel method to study the electrodynamic behavior of actin filaments. Evidence for cable-like
properties of actin. Biophys J 65, 1371–1378, https://doi.org/10.1016/S0006-3495(93)81188-3 (1993).
19. Sataric, M. V., Ilic, D. I., Ralevic, N. & Tuszynski, J. A. A nonlinear model of ionic wave propagation along microtubules. Eur Biophys
J 38, 637–647, https://doi.org/10.1007/s00249-009-0421-5 (2009).
20. Odde, D. Diffusion inside microtubules. Eur Biophys J 27, 514–520 (1998).
21. Shen, C. & Guo, W. Ion Permeability of a Microtubule in Neuron Environment. J Phys Chem Lett 9, 2009–2014, https://doi.
org/10.1021/acs.jpclett.8b00324 (2018).
22. Priel, A. & Tuszynski, J. A. A nonlinear cable-like model of amplified ionic wave propagation along microtubules. Europhysics Letters
83, 68004, https://doi.org/10.1209/0295-5075/83/68004 (2008).
23. Rostovtseva, T. K. & Bezrukov, S. M. VDAC inhibition by tubulin and its physiological implications. Biochim Biophys Acta 1818,
1526–1535, https://doi.org/10.1016/j.bbamem.2011.11.004 (2012).
24. Baskin, T. I. & Gu, Y. Making parallel lines meet: transferring information from microtubules to extracellular matrix. Cell Adh Migr
6, 404–408, https://doi.org/10.4161/cam.21121 (2012).
25. Putnam, A. J., Schultz, K. & Mooney, D. J. Control of microtubule assembly by extracellular matrix and externally applied strain. Am
J Physiol Cell Physiol 280, C556–564, https://doi.org/10.1152/ajpcell.2001.280.3.C556 (2001).

Scientific Reports | (2019) 9:6110 | https://doi.org/10.1038/s41598-019-42343-2 8

Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/ www.nature.com/scientificreports

26. Santelices, I. B. et al. Response to Alternating Electric Fields of Tubulin Dimers and Microtubule Ensembles in Electrolytic Solutions.
Sci Rep 7, 9594, https://doi.org/10.1038/s41598-017-09323-w (2017).
27. Petry, S. & Vale, R. D. Microtubule nucleation at the centrosome and beyond. Nat Cell Biol 17, 1089–1093, https://doi.org/10.1038/
ncb3220 (2015).
28. Starr, D. A. & Fischer, J. A. KASH ‘n Karry: the KASH domain family of cargo-specific cytoskeletal adaptor proteins. Bioessays 27,
1136–1146, https://doi.org/10.1002/bies.20312 (2005).
29. Prat, A. G. & Cantiello, H. F. Nuclear ion channel activity is regulated by actin filaments. Am J Physiol 270, C1532–1543, https://doi.
org/10.1152/ajpcell.1996.270.5.C1532 (1996).
30. Rangan, A. V., Cai, D. & McLaughlin, D. W. Quantifying neuronal network dynamics through coarse-grained event trees. Proc Natl
Acad Sci USA 105, 10990–10995, https://doi.org/10.1073/pnas.0804303105 (2008).
31. Watanabe, H. Coarse-grained information in formal theory of measurement. Measurement 38, 295–302, https://doi.org/10.1016/j.
measurement.2005.09.005 (2005).
32. Lindgren, K. An Information-Theoretic Perspective on Coarse-Graining, Including the Transition from Micro to Macro. Entropy 17,
3332–3351, https://doi.org/10.3390/e17053332 (2015).
33. Doxsey, S. Re-evaluating centrosome function. Nat Rev Mol Cell Biol 2, 688–698, https://doi.org/10.1038/35089575 (2001).
34. Mamon, L. A. Centrosome as “a brain” of an animal cell. Tsitologiia 50, 5–17 (2008).
35. Diviani, D. & Scott, J. D. AKAP signaling complexes at the cytoskeleton. J Cell Sci 114, 1431–1437 (2001).
36. Turnham, R. E. & Scott, J. D. Protein kinase A catalytic subunit isoform PRKACA; History, function and physiology. Gene 577,
101–108, https://doi.org/10.1016/j.gene.2015.11.052 (2016).
37. Pokorny, J., Hasek, J. & Jelinek, F. Electromagnetic field of microtubules: effects on transfer of mass particles and electrons. J Biol
Phys 31, 501–514, https://doi.org/10.1007/s10867-005-1286-1 (2005).
38. Priel, A., Ramos, A. J., Tuszynski, J. A. & Cantiello, H. F. A biopolymer transistor: electrical amplification by microtubules. Biophys
J 90, 4639–4643, https://doi.org/10.1529/biophysj.105.078915 (2006).
39. Schaub, S., Meister, J. & Verkhovsky, A. B. Computational approach to evaluate actin network structure and dynamics based on the
optical microscopy. Molecular Biology of the Cell 13, 194a–194a (2002).
40. Cantiello, H. F., Patenaude, C. & Zaner, K. Osmotically induced electrical signals from actin filaments. Biophys J 59, 1284–1289,
https://doi.org/10.1016/S0006-3495(91)82343-8 (1991).
41. Woolf N. J., P. A. & Tuszynski, J. A. In Nanoneuroscience. Biological and Medical Physics, Biomedical Engineering (Springer, 2009).
42. Nguyen, H. D., Yoshihama, M. & Kenmochi, N. New maximum likelihood estimators for eukaryotic intron evolution. PLoS Comput
Biol 1, e79, https://doi.org/10.1371/journal.pcbi.0010079 (2005).
43. Kawashita, S. Y., Sanson, G. F., Fernandes, O., Zingales, B. & Briones, M. R. Maximum-likelihood divergence date estimates based
on rRNA gene sequences suggest two scenarios of Trypanosoma cruzi intraspecific evolution. Mol Biol Evol 18, 2250–2259, https://
doi.org/10.1093/oxfordjournals.molbev.a003771 (2001).
44. Fisher, R. A. Statistical Methods and Scientific Inference, 2nd ed., (Oliver and Boyd, 1959).
45. Frieden, B. R. & Frieden, B. R. Science from Fisher information: a unification. 2nd edn, (Cambridge University Press, 2004).
46. Frieden, B. R. Physics from Fisher information: a unification. (Cambridge University Press, 1998).
47. Frieden, B. R. Fisher information as the basis for the Schrödinger wave equation. American Journal of Physics 57, 1004–1009, https://
doi.org/10.1119/1.15810 (1989).
48. Frieden, B. R. & Gatenby, R. A. Exploratory data analysis using Fisher information. (Springer, 2007).
49. Shannon, C. E. The mathematical theory of communication. 1963. MD Comput 14, 306–317 (1997).
50. Frieden, B. R. Probability, statistical optics, and data testing: a problem solving approach. 3rd edn, (Springer, 2001).
51. Frieden, B. R. & Gatenby, R. A. Power laws of complex systems from extreme physical information. Phys Rev E 72, https://doi.
org/10.1103/PhysRevE.72.036101 (2005).
52. Gatenby, R. A. & Frieden, B. R. Application of information theory and extreme physical information to carcinogenesis. Cancer
Research 62, 3675–3684 (2002).
53. Kullback, S. & Leibler, R. A. On information and sufficiency. Annals of Mathematical Statistics 22, 79–86, https://doi.org/10.1214/
aoms/1177729694 (1951).
54. Frieden, B. R. & Gatenby, R. A. Information Dynamics in Living Systems: Prokaryotes, Eukaryotes, and Cancer. Plos One 6, https://
doi.org/10.1371/journal.pone.0022085 (2011).
55. Gatenby, R. A. & Frieden, B. R. Information dynamics in carcinogenesis and tumor growth. Mutat Res 568, 259–273, https://doi.
org/10.1016/j.mrfmmm.2004.04.018 (2004).
56. Gatenby, R. & Frieden, B. R. Investigating Information Dynamics in Living Systems through the Structure and Function of Enzymes.
Plos One 11, https://doi.org/10.1371/journal.pone.0154867 (2016).
57. Cunningham, J. et al. Intracellular electric field and pH optimize protein localization and movement. PLoS One 7, e36894, https://
doi.org/10.1371/journal.pone.0036894 (2012).
58. Motlagh, M. S., Khuzani, M. B. & Mitran, P. On Lossy Joint Source-Channel Coding in Energy Harvesting Communication Systems.
Ieee T Commun 63, 4433–4447, https://doi.org/10.1109/Tcomm.2015.2472012 (2015).
59. He, L. D., Han, D. F. & Wang, X. F. Optimal control over a lossy communication network based on linear predictive compensation.
Iet Control Theory A 8, 2297–2304, https://doi.org/10.1049/iet-cta.2014.0322 (2014).
60. Barr, K. C. & Asanovic, K. Energy-aware lossless data compression. Acm T Comput Syst 24, 250–291, https://doi.
org/10.1145/1151690.1151692 (2006).
61. Guppy, M., Kong, S. E., Niu, X., Busfield, S. & Klinken, S. P. Method for measuring a comprehensive energy budget in a proliferating
cell system over multiple cell cycles. J Cell Physiol 170, 1–7, doi:10.1002/(SICI)1097-4652(199701)170:1<1::AID-JCP1>3.0.CO;2-S
(1997).

Acknowledgements
This study was supported by the National Cancer Institute Physical Science Oncology Center Grants U54
CA143970 and by the NCI CCSG Support Grant P30 CA076292.

Author Contributions
B.R.F. developed the information-based models, R.A.G. provided the biological context and hypothesis, both
authors wrote the main manuscript, R.A.G. prepared the figures.

Additional Information
Competing Interests: The authors declare no competing interests.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.

Scientific Reports | (2019) 9:6110 | https://doi.org/10.1038/s41598-019-42343-2 9

Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/ www.nature.com/scientificreports

Open Access This article is licensed under a Creative Commons Attribution 4.0 International
License, which permits use, sharing, adaptation, distribution and reproduction in any medium or
format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre-
ative Commons license, and indicate if changes were made. The images or other third party material in this
article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons license and your intended use is not per-
mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the
copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

© The Author(s) 2019

Scientific Reports | (2019) 9:6110 | https://doi.org/10.1038/s41598-019-42343-2 10

Content courtesy of Springer Nature, terms of use apply. Rights reserved
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:

1. use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
2. use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
3. falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
4. use bots or other automated methods to access the content or redirect messages
5. override any security feature or exclusionary protocol; or
6. share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at

[email protected]