3. A's faulty criticisms only show the strength of the AI Hypothesis

3.1.Fast Na (sodium) ion efflux repeatedly affirmed-against popular opinion
When radioactive sodium ion became available, Levi and Ussing carried out and reported
the historically first sodium-ion-efflux studies---efflux meaning outward flow or movement,
while influx means inward flow. In this study these workers first immersed an isolated frog
sartorius muscle---comprising some one thousand entirely similar long hair-like muscle cells-
-- in a Ringer's solution---a man-made complex salt solution in which isolated cells can be
kept alive for some time--- containing radioactively-labeled sodium ion. They then
proceeded to follow the time course of the loss of radioactivity (from the radioactively
labeled sodium ion) from the muscle as it is being continually washed in a stream of Ringer's
solution containing no radioactivity. The logarithm of the labeled sodium-ion concentration
remaining in the muscle at different times after washing began (as ordinate) was then
plotted against the time of washing (as abscissa). This semilogarithmic plot can be readily
resolved into two "straight-line" fractions, one fast (with a steeper slope) and one slow
(with a flatter slope). Levi and Ussing attributed the fast fraction to come from labeled
sodium ion trapped in the space between the individual muscle cells or fibers (called the
interfibrillar or extracellular space) and the slow fraction as representing labeled sodium ion
emerging from within the muscle cells and referred to as the sodium-ion efflux. This
assignment was soon accepted by almost all the workers in the field. Though not everyone.
I began to have doubts about this assignment in the 1950's, long before I carried out the
energy balance study mentioned above. One strong reason for my doubt was that the size
of the fast fraction of labeled sodium ion leaving the muscle (25%) is much larger than what
one can expect from the known size of the extracellular space (ca. 10%). This discrepancy
led me to suspect that the true sodium-ion efflux from the cells is mixed with that from the
extracellular space, making up a part of the fast fraction. If my suspicion is right, its
implication would be that the conventional value for the efflux rate of sodium is
underestimated by a big factor. This early idea has much to do with the choice of specific
technique for measuring the sodium ion efflux in the energy balance study reported in 1962
(see below). In years following several new lines of approach to this problem have verified
my early suspicion in an unambiguous way. Somet of these new lines of approach are
summarized next:

3.1.1.From single muscle fiber studies:
Early in the 1950's, I put together an apparatus-assembly (the key component is a well-type
gamma-scintillation counter) in which the "detector" encloses the radioactive sample and
thus allows many more readings of the radioactivity emanating from the muscle within a
given period of time than the older methods often used (see below). With a special adaption
of this new apparatus, I studied the sodium-ion efflux from a single isolated muscle cell. The
results showed that even though I was measuring efflux from a single cell (which has no
interfibrillar or extracellular space; the radioactive fluid on the cell surface is washed away
almost instantly and makes no contribution to the much slower "fast fraction") and hence
free from the complication due to the presence of radioactivity trapped in the interfibrillar
or extracellular space of whole muscles, the efflux curve remains curved and---like that from
an intact sartorius muscle containing some one thousand muscle cells--- resolvable into the
two straight-line fractions. This observation confirmed my suspicion that the sodium ion
leaves (and enters) the cell at a much faster rate than widely believed, and is represented by
the fast fraction measured from the single cell. I also gave reason and evidence why the
slow fraction is, in fact, rate-limited by the (slower) rate of desorption of a part of the
sodium ion in the cells which is not free but adsorbed on intra-cellular macromolecules,
mostly proteins.

A new question arose: "How can I reconcile my finding with the earlier report of Sir Alan
Hodgkin and Paul Horowicz (J. Physiol. (London) 148: 127, 1959) who showed that in their
single-muscle-cell experiment, the entire sodium-ion-efflux curve in a semilogarithmic plot
could be fitted to a single straight line? An answer for this apparent conflict was soon
obtained, an answer---- which despite its interesting and intrinsic scientific value, might
have caused fallout affecting me and those closely associated with me far beyond my
understanding of how scientists should conduct themselves when faced with new findings
which run against their prior positions (see Homepage under "Absolute Power Corrupts
Absolutely").

In the technique used by Hodgkin and Horowicz, only a small part of the radioactivity in the
single muscle fiber is "seen" by the detecting Geiger Counter placed at a distance below the
muscle cell sample. As a result, they must wait a longer period of time to collect enough
"counts" for each data point. Accordingly their data points are separated by long 10 minute
intervals. With my more efficient setup I had no difficulty collecting a far larger share of the
radioactivity emanating from the single muscle fibers and was therefore able to measure
data points at 1 minute or even briefer intervals.

Confirming my suspicion that their apparent single straight line relation in the
semilogarithmic plots might be an artifact arising from too few data points, I was able to
show that if I deliberately removed from mmy own data all the data points except those at
the 10 minute intervals--- which was curved and resolvable into two straight lines to begin
with --- the remaining points now also fits a single straight line just like those of Hodgkin and
Horowicz.

The work just described was published originally in 1970 (Physiol. Chem. Phys. 2: 242, 1970).
At that time A was working in my laboratory. Indeed, as indicated above, the key figure
presented in this 1970 paper was reproduced in the 1973 review which A , Ochsenfeld and I
myself coauthored (Ann. New York Acad. Sci., 204, pp. 17-18, Figs. 3 & 4, 1973). I bring this
out here to leave no doubt that A at the time when he wrote his Ph. D. thesis knew that I
have long regarded the conventional assignment (of the slow fraction as representing the
sodium efflux from within the cells) as erroneous and much too slow, and that I believed
that it is part of the fast fraction from the whole muscle which represents the true sodium
ion efflux from within the cells. Three additional sets of investigations further and
unanimously confirmed this early conclusion:

3.1.2 From muscle minus interfibrillar fluid
With a new centrifugation technique Ling and Walton first introduced in 1975 (Physiolo.
Chem.Phys. 7: 215) a simple and quantitative way of removing all the extracellular space
fluid from a frog sartorius muscle became available. Using this method Ling and Walton
removed the labeled sodium ion caught in the extracellular space of labeled-sodium-loaded
muscle before washing began. Despite the quantitative removal of what Levi and Ussing
thought to be the sole source of the fast fraction of the sodium-ion efflux, the
extracellualar-space fluid, a fast fraction (and slow fraction ) remains. This study confirms
what we found under (1) and suggests that a major part of the fast fraction indeed comes
from inside of the cells (Ling and Walton, Physiol. Chem.Phys. 7:501, 1975).

3.1.3. From dying muscles.
When frog muscle is exposed to a Ringer solution containing a low concentration of the
metabolic poison, sodium iodoacetate (IAA), the muscle slowly deteriorates until it dies. In
the course of this dying process, the total sodium ion concentration in the muscle cells rises
slowly from its low value in normal cell ( ca. 25 millimoles per kilogram fresh muscle) to
approach that in the outside Ringer solution ( ca. 100 millimoles per liter or 100 mM.)
Now if to the IAA-containing Ringer solution, we also added radioactively labeled sodium
ion, and expose an isolated frog muscle for some time to this solution (call it solution A),
followed by washing the radioactive isotope-loaded muscle in a Ringer solution containing
only IAA but no radioactivity (call it Solution B), one can obtain a sodium-efflux plot of the
poisoned muscle. If at the end of say one hour of washing in Solution B , one returns the
muscle to the radioactively labeled -IAA containing solution A for a few more minutes, and
then start washing it again in non-radioactive Solution B, one obtains a sodium-efflux plot of
the muscle in a more advanced state of poisoning. The cycle of soaking and washing can be
repeated again and again for a number of times, each time producing yet another efflux plot
of the muscle in a more and more advanced state of poisoning until the muscle dies. All
these plots can be resolved into a slow and a fast fraction. When all the efflux curves are
plotted side by side and compared, one finds that neither the size nor the slope (i.e., rate of
efflux) of the slow fraction changes much (at approximately 20-30 mM.) from the time the
muscle was virtually normal until it was dead---in contradiction to the conventional
assumption that it is the slow fraction which represents the intra-cellular sodium ion.
Otherwise, the size of the slow fraction should gradually rise until it approaches the
concentration of sodium ion in Solution A at around 100 mM.

In contrast, it was the fast fraction which steadily rose in size with each re-immersion in the
labeled solution, and as the cells become closer and closer to death. When the cell dies
completely, the labeled sodium ion in the fast fraction approached that of the sodium ion
concentration in the radioactively labeled soaking Solution A. This experiment established
once more that it is the fast fraction in the centrifuged muscle that represents the sodiumion
efflux. The next set of experiment confirms this conclusion in an even more quantitative
manner (Ling, Walton and Ochsenfeld, J.Cell Physiol. 106:385, 1981).

3.1.4. A rigorous final proof
If one exposes a small frog muscle fiber bundle (containing a few tens of muscle fibers or
cells) for say exactly 3.0 minutes to a Ringer's solution containing labeled sodium ion and
then remove the radioactivity caught in the extracellular space by the centrifugation
method mentioned above, one can then do an efflux curve with closely placed data points.
Each efflux curve obtained can be neatly resolved into a fast fraction and a slow fraction. In
the semilogarithmic plot (done as usual), each fraction again appears as a straight line.
Now each one of these straight lines provides two sets of data to estimate how much
labeled sodium has entered the cells (influx) during the initial 3.0-minute exposure to the
radioactively labeled solution. One set of data comes from the slope of the straight line in
the semilogarithmic plot, representing the rate of efflux (outward flow), but which must
equal the rate of influx (inward flow) since the total Na+ concentration does not change
during the experiment,--- only the (immeasurably small) concentration of radioactively
labeled Na+ in the muscle fiber bundle changes. The other set of data comes from the
extrapolated intercept on the ordinate of the straight line at washing time equal to zero.
The important point is this: only the fraction which truly represents the intracellularextracellular
exchange can yield two values that agree with each other. In contrast, the
fraction which does not represent the true intra-, extra-cellular exchange will produce two
values that disagree.

In a series of studies on 23 small muscle fibers bundles in 8 sets of studies carried out
between Oct. 27, 1977 and Nov. 23, 1977, the ratio of the intake of labeled sodium obtained
from the intercept of the fast fraction over that from the slope of the fast fraction is close to
unity or 100% from the fast fraction (107% + 3.6%) . In contrast, the same ratio from the
slow fraction is far from unity or 100% at 481% + 39.5%. This work provided another set of
unequivocal evidence that the true sodium ion efflux from frog muscle cells is that
represented by the fast fraction (after removal of labeled sodium ion trapped in the
extracellular fluid). (Ling, Physiol. Chem. Phys. 12: 215-232, 1980).

These four sets of independent studies taken together leave no doubt that the
conventional assignment of the slow fraction to represent the sodium ion efflux is wrong.
One consequence of this revolutionary discovery is that the true sodium efflux rate is at
least ten times higher than what has been accepted by most muscle physiologists.
3.2 Affirmation of my earlier "deliberately underestmated" "pumping rate"
With this new understanding in mind, let us now go back to my original energy balance
studies published in "A Physical Theory of the Living State: the Association-Induction
Hypothesis" (Blaisdell, Waltham, 1962, Chapter 8).The first key issue here is how to estimate
the minimum energy need for the postulated sodium pump. Obviously I could not do it the
conventional way by estimating the rate of sodium efflux from the slope of the slow
fraction. However, this was a decision to be made in the 1950's. All the important
experiments to establish unequivocally that it is the fast fraction which represents the
sodium ion efflux still belonged to the future. Nonetheless, I was already very sure that the
slow fraction interpretation is wrong.

So I chose a conservative compromise, which is better than what I knew to be the wrong
answer (i.e., from the slope of the slow fraction) and closer to what was not to be known
with certainty until many years later. I also wanted to choose a compromise that even my
opponents could not legitimately find fault with. Since the majority of workers at that time
believed that the slow fraction represents the entire sodium efflux, I started from there. I
exposed a small muscle fiber bundle for a finite length of time, say 3.0 minutes to a Ringer's
solution containing labeled sodium ion. I then did a standard washout study and resolved
the semilogarimic efflux curve obtained into a fast fraction and a slow fraction. I then
extrapolated the slow fraction---which, I repeat, the majority of muscle physiologists at that
time believed to come from the cells--- to zero time, which would, according to their belief,
yield the labeled sodium ion initially present in the cells of the muscle fiber bundle. Since
that initial radioactively labeled sodium could only have entered the muscle cells during the
3.0 minute of soaking in the labeled solution, dividing that amount of labeled sodium by 3.0
and the muscle-fiber-bundle weight yielded the rate of influx into a unit weight of muscle.
Since during the time of the experiment there was no significant change of the total sodium
ion concentration, the influx rate and efflux rate must be equal at all times. Therefore the
influx estimated from the slow fraction gives me an estimate of the sodium ion efflux rate.
However, I also pointed out clearly that the method I chose for estimating the rate of
sodium ion efflux is not the true initial Na+ but "deliberately underestimated initial Na22
content for the muscle fiber." (Line 6 from bottom of page 209 in the legend of Table 8.7 in
"A Physical Theory of the Living State" 1962, see lp7a), implying clearly that the true efflux
rate is even faster--- as subsequent studies, especially that under (1), (2), (3) and (4) have
established without ambiguity. With these background material fully explained, let us now
turn to Chris Miller's announced reasons given in his Ph.D. thesis for rejecting the AI
Hypothesis and returning to the membrane-pump theory.

3.3. A detailed point-by-point rebuttal of A's criticisms given in his thesis
I put A's criticism into similar 6 subheadings as given by A himself:
3.3.1 Anomalous Influx: (p. 32).
In this, Miller pointed out that the sodium entry I determined by the intercept (i.e., influx
rate) of the slow fraction does not agree with that determined by the slope of the slow
fraction (i.e., efflux rate). This is truly weird.
First note that it was I who introduced the comparison of the efflux rate determined from
the intercepts and from the slopes as a means of deciphering which fraction truly represents
the sodium ion efflux. I was excited about this perception and in the early 70's I talked often
about the intercepts and slopes etc. etc. during our weekly seminar and at other times to all
my graduate and postdoctoral students including A. So much so, he was even a coauthor of
the 1973 review in which the key figure of (1) experimentally establishing this point was
reproduced which showed, like the others, that the slow fraction does not represent the
sodium-ion efflux.
I never said that the efflux rate from the intercept of the slow fraction and from its slope
should agree ---in fact I maintained precisely the opposite: that they should not agree (see
(4) above).
3.3.2.Higher Sodium influx than data from Keynes and Steinhardt. (P. 31).
Hardly surprising. Keynes and Steinhardt's erroneous acceptance of the slow fraction as the
true sodium efflux created their erroneously slow influx values.
3.3.3.Component of passively diffusing fraction left out (p. 26).
Again wrong. It was shown long ago to be negligible compared to the total efflux (e.g., 0.1%)
(Ling, Fed. Proc. Symposium 24: S-103, 1965, p. S-105, footnote).
3.3.4.Large fraction of intracellular sodium not actively transported due to presence in the sarcoplasmic
reticulum (SR).
Anatomically only the T-tubule is directly connected to the outside (Porter and Bonneville,
An Introduction to the Fine Structure of Cells and Tissues, Lea and Febiger, Philadelphia,
1964) and the T-tubule makes up an insignificant fraction of the cell volume,( i.e., 0.4%) to
accommodate the fast fraction of Na+ (Peachy, J. Cell Biol. 25:209, 1965; Hill, J. Physiol.
175:275, 1964) (see Physiol. Chem. Phys. 2: 242-248, 1970, p.246).
The speculation that the fast fraction comes from the SR is not only without supportive
experimental evidence but in fact ruled out by the following two sets of facts considered
together:
3.3.4.1.No fast fraction from K+ efflux
Ling and Walton studied the simultaneous efflux curves of both radioactively labeled Na+
and labeled K+ ion from the same muscles. After incubation of sartorius muscles in a
Ringer's solution containing labeled Na+ ion from the start (total time of incubation in this
radioactive ion was 18 hours at 25o C) but labeled K+ only for the last 25-40 minutes of
incubation, the muscle was centrifuged to remove radioactively labeled ions in the
extracellular space, washed in a stream of non-labeled Ringer's solution and the efflux
curves plotted semilogarithmically as usual. Though freed of labeled ions in the extracellular
space, the Na+ efflux curve continued to show the two-fraction profile as usual, while the K+
efflux is represented by a perfect straight line with a very gentle slope. There is no fast
fraction at all in the K+ efflux curve (Ling and Walton, Physiol. Chem. Phys. 7:501, 1975).
If the fast fraction of Na+ came truly from the SR, then the membrane separating the Ttubule
from the SR proper must be highly permeable to Na+ but impermeable to K+
altogether.
3.3.4.2. Predicted uneven resting potential not observed
In that case, the part of the muscle cell surface near the T-tubule opening would become a
permanent and exaggerated version of the cell surface normally seen only transiently during
the passage of an action potential ( with transient high permeability to Na+ and low
permeability to K+); and as a result, the (standing) resting potential would show great
fluctuation along the length of the muscle fibers, with 90 mV-outside-positive potentials
throughout most of the cell surface disrupted by periodic dips to 0 mV at points near the Ttubules
openings--- which is totally contradicted by facts. No such periodic spatial
fluctuations of resting potentials has been observed. (Ling and Gerard,
J.Cell.Comp.Physiol.,34:383,1949; Nastuk and Hodgkin, J. Cell. Comp. Physiol. 35:39, 1950).
It might also be mentioned that if such spatially sharp difference in potential existed at all, it
would create a wide-spread instability. The cell would be thrown into a continuous state of
disorganized fibrillation and would soon die of exhaustion. Nature, as we know it, is much
wiser than that.
3.3.5.Part of the sodium ion adsorbed, and this reduces the energy need of the pump
This is true. In normal frog muscle, in the presence of an external sodium ion concentration
of 100 mM, about half of the total intracellular concentration is adsorbed. The free fraction
when expressed in micromoles per liter of cell water is about 18 mM. This reduction of free
intracellular sodium ion concentration discovered in my laboratory long after the 1950's
during which the energy balance study was made, would reduce the minimum energy need
somewhat but this reduction is partly compensated by the increase in the sodium ion
concentration gradient which increases the minimum energy need. However, even if we
totally disregard this compensatory effect, the energy discrepancy from the recognition of
the adsorption of intracellular sodium, would lower the minimum energy need of the
sodium pump from between 600 times to 1200 times of the maximally available energy to
between 300 times to 600 times. And that is only the discrepancy from considering just one
ion, the sodium ion, at one kind of membrane, the plasma membrane alone.
3.3.6.In a poisoned muscle, part of the effluxing Na+ may be running down the gradient
This comment has no validity. The turn-over rate of the intra-, extra-cellular exchange of
labeled Na+ goes on at a pace very fast (in minutes) while the time needed to lose the Na+
gradients altogether---never reached in any one of my experiments---would take many
hours if not days. Consider this analogy. If at the end of a day, the water level in your
reservoir has fallen 10%, how much energy you need to keep the water level at its original
level depends not just on the difference of the water level at the beginning and end of the
day, but even more importantly on how fast is the turnover rate of the water. If the
turnover rate is very slow, say many days, then the energy needed to replenish the 10% loss
is just that to move that amount of water. On the other hand, if the turnover rate is ten
times, the energy would be ten times the total water content of the water reservoir. The
10% difference in the initial and final level is of lesser importance. But this is not the only
point where, Miller went astray. He did not study the data carefully.
The intracellular sodium-ion concentration in the poisoned muscle remained essentially
unchanging from its normal initial value of healthy muscles till at least the 6th hour after
poisoning began, hovering between 20 to 30 micromoles per gram of muscle (while the
external sodium ion concentration was at a steady 100 mM). From the 6th hour to the 8th
hour, the sodium ion concentration in the cells rose to about 35 micromoles per gram but
that is still well below the external concentration of 100 mM. Of the three sets of data
presented, two sets lasted only 4 hours during which time there was no gradient
degradation. Only the third set lasted till the 8th hour. But even at the 8th hour in the third
set of experiments the overall electrochemical gradient was still over 100 mV outside
positive (compared to between 120-130 mV outside positive at the very start of the
experiment when the muscles were still normal). It is true that in this particular third set of
experiment, there was such a decline of some 20% of the electrochemical gradient---like the
10% drop in water level of the water reservoir cited above--- and the energy need
accordingly reduced by that percentage---as it was taken into account in my original
calculations. And it was after this correction had been taken into account, that the minimum
need of energy still exceeds the maximum available energy by a ratio of 1800%---a figure
that is vastly made still larger later for reason given above.
To leave no doubt in the mind of the reader, consider another simple thought experiment.
Suppose you are in a row boat on the open sea and suddenly something hit the boat and
water is pouring in. To stay alive, you must bail out the water from the boat fast enough so
that it would stay afloat. If A were right, your job should get easier and easier as the water
level in the boat is rising higher and higher, and more and more of the water in the boat
would leak back to the ocean outside by itself without your help. This is absurd. The truth is
that no water can leak back from a lower level inside the boat to a higher level in the sea on
its own. As long as the boat is still afloat, you must use approximately the same amount of
energy to bail out the same amount of water. To be sure, it would cost a little less energy
due to the lowering of the inside-outside level-difference, but that has already been taken
into account in my calculations.
Similarly, the basic law of physics is incompatible with the idea that a positively charged ion
like sodium could run "down" a 100 mV outside-positive electrochemical gradient, no more
so than water can leak back from inside the still-afloat boat to the ocean outside or water
can flow "down" from the bottom to the top of a hill.
3.3.7. Two other comments.
There remain two other comments made by A in his Ph.D. thesis in defense of the
membrane pump theory and against the AI Hypothesis. One was taken from another pair of
my fleeing students, Palmer and Gulati which is briefly described and answered under linked
page lp7 (6). The other concerning the electrical field effect at the cut end of muscles on K+
was answered in linked page lp7 (1).