ON THE USE OF CARBONIC ANHYDRASE IN CARBONATE AND
`AMINE BUFFERS FOR CO2 EXCHANGE IN MANOMETRIC VES-
`SELS, ATOMIC SUBMARINES, AND INDUSTRIAL CO~. SCRUBBERS
`
`Dean Burk
`National Institutes of Health, Public Health Service, Bahesda, Md.,
`and the Max Planck Institute for Cell Physiology,
`Berlin-Dahlem, German Federal Republic
`
`There is a remarkable resemblance between the kinetics of carbon dioxide
`gas exchange in manometric vessels of I cu. inch volume (rlGIJRE 1)1 and in
`atomic submarines of 100,000 cu. ft. volume (.glGImE 2),2 and it is not the pur-
`pose of this paper to hide such a kinetic similarity. As will be indicated in
`some detail, the over-all mathematical treatment of CO2 in the gas phase is
`much the same whether one is dealing with CO2 absorption in (1) a manometric
`vessel central well (rlOtl~ 1) containing, for example, KOH, NaOH, LiOH,
`K2CO3-KHCO3, Tris buffer (THAM), or mono- or di-ethanolamine; or in (2)
`an atomic submarine CO2 scrubber containing monoethanolamine (rI6tmEs 3
`and 4) or LiOH (FlOtmg 5). The weaker alkaline CO2 buffers mentioned (car-
`bonate, Tris, and the ethanolamines) have all been found to undergo tremendous
`acceleration of CO2 exchange when the enzyme carbonic anhydrase is added; in
`my hands accelerations in rate of 5- to 100-fold have been readily observed
`with all such buffers, depending upon conditions. Strong alkalis (pH > 11 to
`12) destroy the enzyme rapidly, of course.
`The manifold acceleration of C02 exchange by carbonic anhydrase may be
`observed with respect to either absorption or elimination of CO2 gas, depend-
`ing upon the side of the equilibrium point from which one is operating; this
`can be a variable function of, for example, the pH, base exchange, or tempera-
`ture. In this connection it is worth recalling well-known physiological situa-
`tions in which different pH-base equilibria are involved at constant (body)
`temperature. In the stomach mucosal wall and in kidney cells, carbonic an-
`hydrase accelerates the removal of CO2 from the blood, leading to increased
`acidity in the stomach lumen and urine; in the lungs and pancreas, on the other
`hand, carbonic anhydrase accelerates the removal of CO2 gas from the blood,
`leading to more alkaline blood and pancreatic juice. Analogously, in a sub-
`marine CO2 scrubber operated at two different temperatures, carbonic anhy-
`drase could accelerate the absorption of CO2 by the lean buffer mixture in the
`low-temperature absorber and, when this mixture has become enriched with
`CO2 and raised to an appropriately higher temperature in the stripper, carbonic
`anhydrase could accelerate CO2 stripping, provided that the absorbing and/or
`stripping operations were carried out somewhere between 0° and 60° C., the
`practical temperature limits for stable action by this enzyme. Parenthetically,
`it may be remarked that the submarine CO2 scrubber provides an "extra
`breather" for the crew in addition to their lungs. Various engineering possi-
`bilities for feasibility studies in connection with CO2 elimination from atomic
`submarine atmospheres will be considered in the latter part of this paper. Let
`us look at manometric situations first.
`
`372
`
`Akermin, Inc.
`Exhibit 1017
`Page 1
`
`

`

`¢_
`
`FIGURE I. Typical Warburg manometer and illustrative vessel~ for operation at constant
`volume of gas space; x (cu. mm. NTP) gas change = hk, where h -- ram. pressure change
`(I0,000 mm. Brodie fluid = I atmosphere), and k ffi ([Va. 273/T] -{- VLa)/Po, Va -- volume
`of gas phase (cu. ram.), VL = volume of liquid phase (cu. mm.), Po is a pressure of one at-
`mosphere expressed in millimeters of confining fluid, T is the absolute experimental temper-
`ature, and a is the Bunsen absorption coefficient of the gas exchanged.
`
`ATMOSPHERE
`RE*,CTOR COMI~T.-~ ANALYZER7/- OFF. O~RS.
`
`,
`,-HzlCO BURNERS
`CREWS QTRS;’-~y" ’ENO RM
`
`r.~ STERN RM~ ~ ~:. ¯ ~
`
`--
`
`....
`’t’/~~BBERS
`
`/ATTACKCENTE.RT// ELECTROSTATIC
`/ / F PRECIPITATOR
`/
`~-’TORPEDO RM.
`
`~ !~ ~!i~i: ~ /
`’ ",, ""
`"~ OXYGEN FLASKS
`
`~
`MAIN CARBON
`FILTER
`
`FIGURE 2. Atmosphere control equipment on a typical nuclear-powered submarine, with
`special reference to indicated size and location of CO2 scrubbers3
`
`/---CREW~ MESS
`
`373
`
`Akermin, Inc.
`Exhibit 1017
`Page 2
`
`

`

`374
`
`Annals New York Academy of Sciences
`
`Manometric Measurements
`
`The advantages and disadvantages of various ethanolamines as COs buffers
`and absorbents in manometric measurements were pretty well outlined about
`10 years ago by Pardee,* Krebs,4’~ and Burk et al.6 Among such amines for
`manometry, diethanolamine was found to be the most generally suitable with
`respect to desired Kb value, solubility, nonvolatility, COs retention, and main-
`
`FIGURE 3. Submarine CO2 removal plant, front and left-side view, at man-sized scale3
`
`tenance of constant COs gas pressure up to the order of 2 to 3 per cent-atmos-
`phere. The chief disadvantages of even the best of the amines tested were
`residual autoxidation that could not be entirely eliminated, even with 0.1 per
`cent thiourea; diificulty in satisfactorily attaining a physiological COs partial
`pressure of about 5 per cent in the gas phase at about 1 atm. total pressure; and
`rates of absorption considerably lower than those shown by strong alkalis (such
`as KOH and NaOH) in small absorption containers.
`The most interesting observation that can be made about the manometric
`use of ethanolamine COs absorbents at present is that quite recently it has been
`
`Akermin, Inc.
`Exhibit 1017
`Page 3
`
`

`

`Burk: Use of Carbonic Anhydrase in COs Exchange 375
`
`found that they are equaled or surpassed in virtually all aspects by concentrated
`potassium carbonate-bicarbonate buffers containing carbonic anhydrase (War-
`burg and Krippahl,7 and Stambuk and BurkS). The carbonate-bicarbonate
`
`FIGURE 4. Submarine CO2 scrubber, exposed view front and right side. Hardware
`rather than CO~ absorbents occupies most of the space3
`
`buffers can be adjusted readily to provide fixed CO2 pressures over a several
`thousandfold range from less than 0.01 per cent-arm, up to well above 10 per
`cent-atm., with high CO2 retention, no autoxidation, and a satisfactory equilibra-
`tion time of but a few minutes at most, providing that suitable manometric
`vessels are employed, notably and preferably the new vessels described by War-
`
`Akermin, Inc.
`Exhibit 1017
`Page 4
`
`

`

`376
`
`Annals New York Academy of Scienees
`
`burg and Krippahl.9 In such vessels, illustrated in FIGURE 6, the small central
`well on the floor of the main compartment (of the vessel shown in ~’IGURE 1) is
`replaced by an elevated wide-area trough that is optionally further connected
`directly with a side arm; indeed there may be an additional independent side
`arm2 For high physiological CO2 pressures of the orderof 5 per cent-atm., the
`carbonate-bicarbonate buffers will be of the order of 1 to 3 M (and largely bi-
`
`FmtrR~ 5. CO2-canister with six LiOH units capacity3
`
`carbonate), and will contain 5 to 10 Meldrum and Roughton units~° of carbonic
`anhydrase per cc. (that is, about 0.001 rag. highly purified enzyme!cc., or up to
`1 rag. commercially prepared enzyme!cc., depending upon the purity of the
`latter). The vessels with a trough to contain buffer also improve the usage
`of the amine buffers by similarly decreasing the equilibration time, as a result
`of increased area and volume of buffer that may be employed conveniently in
`a given vessel.
`The new manometric vessels, carbonate-bicarbonate buffers, and carbonic
`
`Akermin, Inc.
`Exhibit 1017
`Page 5
`
`

`

`Burk: Use of Carbonic Anhydrase in COs Exchange 377
`
`anhydrase thus provide, at long last, a one-vessel method for measuring at con-
`stant and physiological CO2 gas pressure either the oxygen consumption or oxy-
`gen production by animal or plant cells or tissues suspended in a medium and
`pH of choice, and in a medium physically separated from the buffer mixture;
`under such conditions, therefore, the manometers may then register pressure
`changes solely due to oxygen gas exchange uncomplicated by simultaneous
`carbon dioxide gas exchanges, at least under steady-state conditions that we
`shall now consider in some detail. Here it is important to distinguish between
`
`i~It3Lrl~ 6. Manometer vessel with central elevated trough and confluent side-arm,g
`
`two quite different kinds of CO2 steady-state pressures, one determined chiefly
`by buffer equilibria and the other determined chiefly by the rate of CO2 absorp-
`tion in the COs-absorbing liquid. The latter case will be considered first; for it,
`Dixon and Elliottn give a somewhat similar kinetic treatment and derivation
`that offers a useful comparison with that given here, but their results are ex-
`pressed in terms of quantities (volumes) of CO, rather than of pressures of CO,
`that are of prime interest here.
`Case 1. Absorption rate steady-states. The steady-state pressure of CO2 in
`the vessel gas phase, pCO~,, will here be a function of the difference between
`(1) the rate of production of COs by, for example, cellular material in the liquid
`medium in the main compartment of the vessel; and (2) the rate of absorption
`of COs in the central well or trough containing either strong alkalis (KOH,
`
`Akermin, Inc.
`Exhibit 1017
`Page 6
`
`

`

`378
`
`Annals New York Academy of Sciences
`
`NaOH) or relatively alkaline amines or carbonates (low proportion of bicar-
`bonate). Indeed pCO2~ can be shown to be simply RVL/a, where R is the rate
`of production of CO2 expressed as cu. mm. normal temperature and pressure
`(NTP)/unit t/cc. liquid medium in the main compartment, VL is the number
`of such cc. in the compartment, t is time, and a is the specific absorption rate
`constant whose numerical and algebraic values (see below) depend upon factors
`such as the exposed cross-section area of the buffer in the well or trough, the
`kind and concentration of CO2 absorbent in the well or trough, the rate of
`shaking, the temperature, and the geometry of the vessel. However, a can be
`determined empirically, for any given set of conditions and vessel to be em-
`ployed, by observing the change in pressure with time when CO2 gas, which has
`been liberated suddenly from a medium containing bicarbonate (or carbonate)
`by addition of excess HCI or other strong acid from a vessel side arm, is ab-
`sorbed by alkali or buffer in the well or trough; when the natural logarithm of
`the pressure is plotted against time, a linear (first order) curve is obtained whose
`slope is a/k, from which a may be obtained by elimination of k, the ordinary
`(and ordinarily predetermined) Warburg vessel constant expressed in desired
`units.* In order for the pressure of CO2 to be expressed in the desired terms of
`per cent-atmospheres NTP (100 per cent = 760 mm. Hg = 10,000 mm. Brodie
`fluid), k is calculated as [10] [(273 VolT + (VL aco~)] sq. ram., where Vo is
`the volume of gas space in the vessel in cc., T is the absolute temperature, and
`aco~ is the Bunsen solubility coefficient of CO~ in the liquid medium. The factor,
`k, converts a CO~ gas pressure change, hco,, in per cent atmospheres (per cent-
`arm.) into cu. ram. NTP CO~ gas change in the vessel (see legend to rIGt~RE
`1), and is here exactly 100 times the numerical value of the ordinary Warburg
`vessel constant, k, that converts CO~ gas pressure change in mm. Brodie fluid
`to cu. ram. NTP CO2 gas change in the vessel.
`At the start of any manometric experiment, the steady-state pressure of CO~
`in the vessel gas phase is not established instantaneously, but will usually
`require some minutes or seconds for practical attainment, in accordance with
`the equation
`
`(1)
`
`* The rate of absorption of CO, in the well or trough can readily be shown experimentally
`to be proportional at any moment to the pressure of CO~ in the gas phase; that is,
`
`-d pCO~/dt = (a/k) pco~,
`
`d pCO~/pCO~
`
`~
`
`(a/k)
`
`ft~
`
`dr,
`
`f pC02
`
`J~co2’
`
`In (pCO2/pCO2’) = (a[k) l,
`
`(A)
`
`(B)
`
`(C)
`
`and
`
`where pCO~ is the pressure observed at time t and pCO~~ is any subsequent pressure at sub-
`sequent time f, and a/k is the slope of Equation C.
`
`Akermin, Inc.
`Exhibit 1017
`Page 7
`
`

`

`Burk: Use of Carbonic Anhydrase in COs Exchange 379
`
`This equation yields, upon integration, and evaluation of the integration con-
`stant by placing pCO2 = 0 when t = 0,
`
`Placing
`
`pCO.~ = (RVL/a)(1 -- e-a’/k).
`
`pC02~, = RVL/a
`
`(2)
`
`(3)
`
`It is evident from Equation 3 that, once the necessary constants have been
`predetermined for a given set of conditions, calculation of the steady-state
`pressure of CO,, in the vessel gas phase is very simple and exact, as are like-
`wise, from Equation 2, the times required to attain particular fractions of the
`steady-state pressure. Thus the time required to attain 90 per cent of the
`steady-state value may be calculated by putting pCO2 = 0.9 pCO2, in Equa-
`tion 2, whence
`
`0.9 (RV,~/a)
`
`and
`
`so that
`
`-at/k = In 0.1,
`
`t90% = -- (2.3 log O.1)/(k/a).
`
`In an ordinary manometric vessel similar to that illustrated in FIGURE 1, con-
`taining 2 N KOH in the well, k is usually about 150 ram.2 for a vessel of some
`18 cc. volume, and a is commonly about 70 min.-1, whence the time required
`for attaining 90 per cent pCO2, is about 5 min. and, for attaining 99 per cent
`pCO~, about 10 min. (twice as long, precisely). It is to be noted that the time
`for any fractional attainment of CO2 steady-state pressure is independent of
`the rate of CO~ production, RVL. On the other hand, a steady-state pressure
`of CO2 in the vessel gas phase of 0.1 per cent-arm. (= 10 mm. Brodie fluid)
`would occur, with a about 70 min.-1, when RVL/a = 0.1, or RV~ = 7 cu. mm.
`CO2/min., or h = about 5 ram. pressure change of CO2 per minute.
`Obviously, the steady-state pressure of CO2 in the vessel gas phase can be
`varied by any of the ways in which R, V,~, and a can be varied, as best suits
`the investigator with the particular apparatus he has at hand. In general, the
`steady-state pressure will be lower the more effective is the COs absorption (the
`greater the value of a), the smaller the value of V,., the smaller the concentra-
`tion of cellular matter, or the lower the rate of CO2 production per unit of cel-
`lular material. With appropriate choice of a, V,., and (especially) R, it is
`readily possible to attain and to maintain steady-state pressures of CO~ as low
`as 0.001 per cent-atm. (0.1 ram. Brodie fluid). R may be varied widely by em-
`ploying widely varying concentrations or amounts of cellular material per ves-
`sel. Such a methodology allows one to study the effect of very low concentra-
`tions of CO~ upon various metabolic processes, such as the respiration or growth
`of microorganisms. Years ago, co-workers and I were able to show~ that the
`
`Akermin, Inc.
`Exhibit 1017
`Page 8
`
`

`

`380
`
`Annals New York Academy of Sciences
`
`growth of a given strain of yeast, Saccharomyces cerevisiae strain Fleischmann
`139,13 was still maximal at a CO~ pressure of no greater than 0.001 per cent-arm.
`when the cells were grown with biotin; when the cells were grown with the di-
`aminocarboxylic (opened-ring) derivative of biotin (DAC), however, CO2 pres-
`sures about 100 times as great were required; likewise, at 1 per cent-arm. CO~,
`the growth versus biotin concentration curve attained its half-maximum and
`maximum values at concentrations of biotin that were 100 times smaller than
`those of DAC for the corresponding growth versus the DAC concentration
`curve and, at 0.001 per cent-atm. CO,, this difference was increased another
`100-fold (10,000-fold altogether). Such data were part of our evidence at the
`time14 that the function of biotin in living organisms was closely connected with
`the CO~ requiremnet, a view that has been fully substantiated since by many
`workers, especially by the recently added chemical evidence provided in various
`papers by Lynen and his co-workers.
`In TABLE 1 is given illustrative, previously unpublished data taken by us 30
`years ago on the effect of CO~ pressure on the growth velocity* of the nitrogen-
`fixing bacterium, Azotobacter vinelandii, that is an application of material pre-
`sented in the two previous paragraphs; here, since growth over a considerable
`period of time was involved (7 hours), there was some shifting in the steady-
`state pressures that required a certain amount of averaging with respect to
`time, but without notable distortion of essential end results and conclusions.
`It is to be noted that the growth was logarithmic, and that the velocity con-
`stants of growth were calculated from semilog plots of either respiration rate or
`turbidity with respect to time.
`
`* It was established more than 30 years ago that a minimum pressure of carbon dioxide is
`essential for the normal growth of many, if not all, heterotrophic (as well as autotrophic) or-
`ganisms. An absolute CO~ requirement was shown to exist clearly distinguishable from
`beneficial stimulation, utilization as a source of carbon, or buffering action in the external
`medium. Such findings were established by experimental methods that were quahtatively
`accurate but, for the most part, semiquantitative at best. In general, cultures were grown in
`the presence of Cot-absorbing alkali without, however, the concentrations of COt in gas or
`liquid (or solid) phases being adequately defined. Occasionally, streams of CO~-free and CO2-
`containing air or nitrogen were employed, but again with the same disadvantage.
`Even when cultures were grown with large but confined gas volumes, the initial pressures
`of COt became altered during the course of the experiment by additional metabolic COt
`formation; satisfactory corrections therefore were difficult to apply since the CO2 require-
`ments of most organisms are satisfied at very low pressures of CO~, usually below those existing
`in air (0.03 per cent-atm.), and any small amounts of COt that were produced created relatively
`large pressure changes with respect to CO~ pressure requirements. Most of the experiments
`were carried out with stationary or slightly shaken cultures, with the result that it was not
`possible to determine, even by thermodynamic calculation, the relative or absolute amounts of
`C03- -, HCOs-, H~COs, or COt in the liquid phase from the COt in the gas phase, even when
`the latter was known approximately. This is an important consideration if it is desired to
`further determine which of these substances might be mechanistically responsible for the "COS
`effect."
`Although equilibrium between the three first-named forms is established almost instan-
`taneously, the formation of C02 from them, or vice versa, is relatively slow and of the same
`general order of rate as the still further exchange of COs between gas and liquid phase.
`Finally, the growth function in the early experiments was not based on studies of the
`velocity constant of growth, but only upon the amount of growth; the latter, in contrast to the
`former, necessarily varies with the duration of the experiment. TABIm 1 represents a type of
`experimentation in which many of these problems and difficulties are overcome in good meas-
`ure. Space considerations and relative interest do not permit presentation here of consid-
`erations of COa- -, HCOa-, and H,COa functions that may be derived from the experimenta-
`tation illustration in TABLE 1, but many interesting kinetic and thermodynamic deductions
`could be drawn from such experimentation.
`
`Akermin, Inc.
`Exhibit 1017
`Page 9
`
`

`

`Burk: Use of Carbonic Anhydrase in CO2 Exchange 381
`
`Case 1 continued. Absorption rates with different types of manometric vessds
`and CO,z absorbents. Before considering Case 2 it will be desirable to give a
`brief historical sketch to date of actually observed rates of CO2 absorption in
`various types of manometric vessels in common use with the principal types of
`CO2 absorbents of interest here. The Barcroft manometric vessel illustrated in
`:FIGURE 7, developed in England many decades ago, is an example of a vessel
`
`THE VELOCITY CONSTANT OF GROWTH OF AZOTOBACTER AS A FUNCTION OF
`PRESSURE OF CO.z IN TH-E GAS PHASE*
`
`TABLE 1
`
`VL, culture per vessel (cc.)
`Va, gas space per vessel (cc.)
`k, vessel constant (ram.2)
`a, specific absorption constant (hr.-1)
`RVL, respiration rate per vessel (cu.
`ram. O~ or CO2/hr.; R.Q. = 1)
`Initial (t -- 0 hr.)
`Final (t = 7 hr.)
`Meaner (per hour)
`pCO2~ (mean per cent-arm. CO, in
`gas phase [RVL/a])
`Final growth~ (t --- 7 hr.)
`Respiration rate (cu. ram. O2/
`hr./cc.)
`Relative turbidity (per cc.)
`g,§ mean velocity constant of growth
`From respiration rate
`From relative turbidity
`Average
`
`0.5
`17.01
`153
`2700
`
`1.0
`18.06
`162
`2700
`
`2.0
`18.48
`166
`3350
`
`3.0
`17.98
`161
`3350
`
`4.0
`17.98
`161
`3350
`
`5.0
`9.0
`6.7
`
`10.0
`26.0
`16.0
`
`20.0
`64.0
`36.0
`
`30.0
`102.0
`56.0
`
`40.0
`144.0
`76.0
`
`0.002!
`
`0.0059
`
`0.0108
`
`0.0168
`
`0.0227
`
`18
`54
`
`26
`84
`
`32
`100
`
`34
`108
`
`36
`114
`
`0.084
`0.076
`0.080
`
`0.139
`0.139
`0.139
`
`0.168
`0.165
`0.167
`
`0.177
`0.176
`0.177
`
`0.188
`0.188
`0.182
`
`The material used was a 1-day-old Azolobacter z4nelandii culture, at 31°C. temperature,
`pH of 6.8, in an inorganic medium with 1 per cent glucose. The duration of the experiment
`was 7 hours. The source of nitrogen for growth was the N2 gas in air; the vessel was of the
`type illustrated in rmv~.~ 1, containing 0.3 cc. of 2 N NaOH in the central well. The CO~
`in the gas phase was supplied by that formed in respiration. The total pressure in the gas
`phase was 1 arm., and the rate of respiration per unit of cells was constant with time, so that
`an increase in rate per vessel with time is a measure of the rate of growth and is accompanied
`by a corresponding increase in the number of organisms, turbidity, cell nitrogen, and other
`factors. The manometers can be read as frequently as desired; hence, time curves of the
`increase in growth may be determined over relatively short periods of time, such as hours or
`minutes.
`* These previously unpublished experiments were carried out with Hans Lineweaver and
`C. Kennetli Horner in 1930 at the Fixed Nitrogen Research Laboratory, United States
`Department of Agriculture, Washington, D.C.
`~ Antilog ([log final rate ÷ log initial rate]/2). Since the increase in rate of respiration
`due to growth was logarithmic and not linear with respect to time, the true average rate is
`given by the logarithmic mean and not by the arithmetic median; that is, not by (initial
`rate + final rate)/2. The arithmetic median values are, however, only slightly higher than
`the lgoarithmic mean values, namely, 6.75, 18, 42, 66, and 92 respectively.
`:~ The initial "growth" at t = 0 was, per cc. of culture: respiration rate, 10 cu. mm. O~ihr.;
`relative turbidity, 32.
`§The velocity constant of growth, g, is given by 2.3 (log final growth -- log
`initial growth)/t, since growth is (and was) a logarithmic (first order) process, the velocity
`of which is proportional at any moment to the concentration of cells (that is, to respiration
`rate, relative turbidity, and other factors) obtaining at that moment. In the present ex-
`periments, the rates of respiration for the different vessels were measured every hour; upon
`plotting these rates semilogarithmically against time, straight lines were obtained that re-
`vealed a very slight bending upward corresponding to the increases in partial pressure of
`C02 in the gas phase as the rates themselves increased (that is, as RVL/a values increased).
`The greatest relative increases took place where the increase in CO~ pressure had the least
`effect: in the range where the limiting maximum growth velocity was being approached.
`
`Akermin, Inc.
`Exhibit 1017
`Page 10
`
`

`

`382
`
`Annals New York Academy of Sciences
`
`that is "practically useless’’n for rapid absorption of CO2 even by strong alkali
`in the central well unless filter paper wetted with the alkali is placed in the well
`(FIGURE 8). The area of the exposed liquid in the well without filter paper is
`too small in relation to the volume of gas space involved (about 35 cc.), so that
`without the filter paper one-half hour or much more is required to absorb a rea-
`sonably small amount of CO2 in the gas phase, with strong alkali as absorbent
`and, of course, longer with the CO2 buffer absorbents. Likewise, as indicated
`in FIGURE 9, when a manometric measurement of respiration is taking place,
`
`FIGURE 7. Side view of Barcroft manometric vessel attached to differential manometer,n
`one side only shown.
`
`too long a time is required for the building up of the steady-state CO2 pressure;
`correspondingly, the steady-state pressure of CO~ is relatively quite high as a
`consequence. Thus in FIGURE 9, the 84 cu. mm. COs in the gas phase corre-
`sponds to about 30 mm. Brodie pressure of COs at the steady-state, or about
`about 0.3 per cent-atm., as compared to 0.1 per cent-atm, or less mentioned
`earlier in connection with a vessel of the type shown in FIGURE 1, with both
`types of vessels here compared without use of the filter paper device. For
`comparable respiration rates per volume of vessel, the steady-state pressure of
`CO~ would be about six times as high in the vessel of ~’IGUI~E 7 compared to
`the half-as-large vessel of FIXTURE 1. Vessels of the type shown in FIXTURE 7 are
`
`Akermin, Inc.
`Exhibit 1017
`Page 11
`
`

`

`Burk: Use of Carbonic Anhydrase in CO2 Exchange 383
`
`still much used in that quaint isle to the east of Ireland, but more from patri-
`otic than scientific considerations.
`In any event, none of the well-vessels containing strong alkali as the CO.,.
`absorbent permits establishment of physiological COs steady-state pressures of
`200 ~,
`
`"~
`
`~L 40’/oKOI"
`
`~-~.D ~ FILITER P~APER
`
`-50 ~’C J ~IN W~LL
`
`F]GU~ 8. Rate of absorption of CO~ in Barcroft vessel under different e~erimental
`conditions.~ Curve A: ~per cent NaOH in central well, temperature 14° C. (curve for ~o C
`shghtly lower); Curve B: ~ per cent KOH in central well, temperature ~o C.; Curves C and
`D: respectively 40 per cent and 7 per cent KOH in central wells, and al~ rolls of filter paper in
`wells on right-hand (experimental) side. Values below ordinate of zero, in curves C and D,
`caused by o~dation of alkafine filter paper by O2, and would not have been observed with
`Whatman filter papers ~ and 42, placed in both right- and left-hand flasks.
`
`I00
`
`|
`~ oo
`
`300 Cll. MM./HR.
`
`o
`o 20
`
`40 6
`Minutes
`
`80 1 oo
`
`FIGURE 9. Curve showing accumulation of CO2 in the gas phase with absorption condi-
`tions=1 corresponding to curve B of I~IG~ 8 (q.v.), and with a CO2 production of 300 cu. mm.
`per hour by the yeast cells in the medium in the main compartment. Length of arrow indi-
`cates value obtained when tap closed off 10 min. after ~ = 0.
`
`Akermin, Inc.
`Exhibit 1017
`Page 12
`
`

`

`384
`
`Annals New York Academy of Sciences
`
`about 5 per cent-atm, for measurement of respiration. To overcome this situa-
`tion, Pardee~ in 1949 introduced the use of diethanolamine in place of strong
`alkali, but with only partial success, since (1) about 3 per cent-arm. CO., was
`the practical maximum steady-state pressure feasible in any type of manomelric
`vessel, and (2) the time required to reach such a steady-state was still at least
`10 min. and usually more even in vessels such as that in FI(;ugr~ 1, although
`this factor was improved upon by use of Dickens-Simer vessels that provide a
`considerably greater area of CO~-absorbing diethanolamine exposed to the gas
`phase (rI6VRE 10). The studies of Krebs4,5 that followed soon afterward are
`to be recommended as a scholarly consideration of the theory and practice of
`use of various ethanolamine buffers as CO~ absorbents in manometry, with
`respect to both advantages and limitations; gentle warning may be given, how-
`ever, that the treatment of "retention" is unnecessarily complicated and is
`greatly simplified in the treatment of retention given by Warburg and Krip-
`pahl.7 The vessel types shown in ~’x6ugE 69 and in ~’~ovgE 115 offer a practical
`solution to the problem of the not-quite-adequate equilibration times met with
`in use of ethanolamines (or carbonates), as compared to strong alkalis in more
`conventional vessel types. The new vessel type in rIGVRE 6 is, in my experi-
`ence, the most generally useful and satisfactory type of manometric vessel
`extant; with a second independent side arm, it will accomplish almost all the
`special functions of the Dickens-Simer and Dixon-Keilin (Summerson) vessels,
`even for use in differential manometry. The equilibration times attained with
`the elevated central trough are remarkably short, as shown in Abb. 3, p. 435 of
`Warburg and Krippahl.9 TABLZS 2 and 3 illustrate the use of the vessels in
`rmvg~ 11~ in measuring respiration and photosynthesis with short equilibra-
`tion or transition times; xA~ 2 compares diethanolamine with NaOH, and
`TABLE 3 shows the sharp transition times obtainable with diethanolamine as
`one proceeds cyclicly from respiration to photosynthesis and back.
`It may be gathered that what is desired in manometry is an equilibrated
`steady-state that is attained within a few minutes or less. Naturally there is a
`limit of time within which a gas may be shaken in or out of the liquid phase at
`feasible rates of gas and liquid shaking in the vessels, and with the least inter-
`ference due to geometric obstruction (side arms, wells, and troughs). Some
`idea of such a practical limit is given by rrCURES 12 and 13, carried out with
`respect to O~ absorption in a "rectangular vessel" without the indicated geo-
`metric obstructions; the vessel containing actinometric fluid was illuminated in
`unshaken condition for 3 min.; during this time about 70 cu. ram. O~ in the
`liquid were consumed, and then, in the dark, was shaken at varying speeds to
`permit re-establishment of equilibrium O~ solubility in the liquid from the gas
`phase. It is seen that a limiting half time of 5 to 8 sec. is possible for practical
`attainment, corresponding to virtually complete absorption within about 40
`sec. at the highest rate of shaking employed, namely at 240 horizontal cycles
`per minute at an amplitude of 2 cm.; between 135 and 120 cycles there is a
`sharp fall-off in half time. The case of O~ is given prior to that of CO~ since,
`in this example with O~ only physical considerations are involved, all the chemi-
`cal reactions of O~ having ceased the instant the light was turned off; whereas
`with CO~ there is always the possibility that known or unknown chemical re-
`actions may be additionally involved.
`
`Akermin, Inc.
`Exhibit 1017
`Page 13
`
`

`

`5
`
`4
`
`0
`0
`
`DICKENS-SIMER
`FLAS KS
`
`AGIDIFI
`
`¯
`~20 cu.mm. CO~
`
`L
`
`%C0,
`
`oo
`
`, .o 0.7
`
`2
`ML. HGL
`
`4
`
`0
`
`2 0
`
`40
`MINUTES
`
`~0
`
`80 NoOH
`
`FIGu~ 10. (Left.) Dependence of COs in gas phase on composition of Pardee solutions in central well, as
`determination by amount of HCI added to diethanolamine-carbonate buffer mixture. (Right.) Rate of uptake
`of 320 cu. mm. COs released at 43 rain., after equilibrium COs pressures had been established above various Par-
`dee mixtures or NaOH. Even in the Dickens-Simer manometric flasks employed, the equilibration and times
`are not quite satisfactory, and vessels of the type shown in FmUR~. 6 and 11 are much better.
`
`

`

`FIGURE 11. Two-compartment vessel for measuring with rapid equilibration the pressure
`changes of one gas (for example, CO2 or O~) while pressure of the other is kept constant (for
`example, O~ or CO2 respectively), by respectively oxy-bis(cobaltodiarnines) or by diethanol-
`amine (or alkalis) placed in the reagent compartment R, with tissue or cells in compartment S.
`(a) Top view; (b) end view; and (c) side view. Length (internal dimensions), 35 mm.; width,
`28 mrn.; height, 38 mm. (or, in another model, 20 mm., with top 18 mm. part eliminated).~
`
`COMPARISON OF RESPIRATION OF CHLOR£ZLA AS MEASURED WITH DIET~IANOLAMINE AND
`
`NAOH AT CO2 PRESSURES OF 2.0 AND 0.01 PER CENT-ATM. RESPECTIVELY
`
`TABLE 2*
`
`Vessel type
`CO2 absorbent
`Compartment
`Vessel volume (cc.)
`ko2, vessel constant (rnrn.2)
`Cu. mm. cell/cc, medium (compartment S or main)
`Cu. mm. cells/vessel
`Pressure change in dark, 30 min. (mm. Brodie)
`(~tuorescent lighting, 30 rain. + 5 min. dark re-
`equilibration, then) :
`Pressure change in dark, 30 rnin. (rnrn. Brodie)
`Total pressure change, 60 min. (mm. Brodie)
`Cu. mm. O~/60 rain.
`Cu. mm. 0~/60 min./lO0 cu. ram. cells (= respira-
`tion rate as a percentage of cu. rnm. O:/cu. turn.
`cells/hr.)
`
`FIGURE II
`Diethanolarnine~
`R
`51.0
`3.85