`
`Proceedings of the
`
`NATIONAL ACADEMY OF SCIENCES
`
`
`Volume /16'
`
`- Number5
`
`- May 15, 1.960
`
`THE BIOLOGICAL FUNCTION OF VITAMIN A ACID
`
`JOHN E. DOWLING AND GEORGE WALD*
`
`BIOLOGICAL LABORATORIES OF HARVARD UNIVERSITY, CAMBRIDGE
`
`Communicated March 24, .1960
`
`The first symptom of vitamin A deficiency in man and other animals is the rise
`of visual threshold known as night-blindness. This is also the only symptom
`the cause of which is well understood. The photosensitive pigments of the rods
`and cones (rhodopsin, iodopsin) upon which the visual threshold depends are formed
`from vitamin A (C19H27CH2OH), through the combination of its aldehyde, retinene
`(C19H27CHO), with specific proteins ofvthe rods and cones called opsins} In rats
`maintained on a vitamin A-deficient diet, after initial stores of vitamin A in the
`liver and blood have been exhausted, the level of rhodopsin in the retina begins to
`fall, the logarithm of the visual threshold reciprocally rising, marking the beginning
`of night-blindness.’
`Several weeks later, the level of opsin in the retina also begins to decline, and
`with this the rod outer segments deteriorate anatomically? An explanation has
`been suggested for this effect also. Opsin in aqueous solution is a much less stable
`protein than rhodopsin.
`It is rapidly denatured by exposures to heat,“ acids and
`alka1ies4 that leave rhodopsin intact. When, as the result of the deficient diet,
`the retina comes to contain opsin that can find no vitamin A with which to com-
`bine, this intrinsic instability is probably the cause of its disintegration; and since
`the outer segments of the rods are largely composed of this protein, as opsin is
`lost their anatomical structure must suffer accordingly.
`The deterioration of opsin and of the rod outer segments, however, is only a
`detail in a much Wider complex of changes occurring throughout the animal at the
`same time;
`for at this time all the overt signs of vitamin A deficiency appear:
`loss of Weight, postural imbalance, respiratory disturbances, corneal opacities,
`disarrangement of coat, and red secretions about the eyes. Tissues, perhaps par-
`ticularly epithelia, have begun to disintegrate in many parts of the body, and
`within a few days more all these animals are moribund.’
`It has long been recognized that vitamin A plays some general role in the tissues,
`indispensable for their integrity and the growth and maintenance of the animal.
`The nature of this, by far its most important function, is as yet altogether unknown.
`Our experiments, having begun with well-understood processes in the retina, had at
`this point become involved with these Wider and wholly obscure phenomena.
`It seemed possible that we might be able to disentangle this situation, and per-
`haps learn something of the tissue function of vitamin A, with the help of vitamin
`587
`
`
`
`588
`
`BIOCHEMISTRY: DOWLING AND WALD
`
`PROC. N. A. S.
`
`A acid. This substance (C19H27COOH), first prepared by Arens and Van Dorp,5
`was shown to maintain growth in the rat and to stave off obvious signs of defi-
`ciency, with a bipotency approaching that of vitamin A itself;“v 7 yet no matter
`how large the amounts in which it was fed, no vitamin A was deposited in the liver.”
`The rat seems unable to reduce vitamin A acid to vitamin A; and this special
`circumstance led Moore9 to suggest that though the tissue functions of vitamin A
`seem to be fulfilled by vitamin A acid, this substance might not be able to serve as
`precursor of the visual pigments, which need for their synthesis the alcohol and
`aldehyde. These considerations formed the starting-point of the present investi-
`gation.
`Methods.——Most of our procedures were described in an earlier paper? Male,
`weanling rats from the Harvard colony were raised on the standard U. S. P. vitamin
`A-deficient test diet, to which supplements were added as wanted. Techniques for
`evaluating liver and blood vitamin A, rhodopsin, and opsin in the eye, and record-
`ing electroretinograms were as described earlier.
`Vitamin A and vitamin A acid, dissolved in cottonseed oil, were administered by
`mouth, through a syringe with a blunted point. Since in these experiments we
`were interested only in maximal effects of vitamin A acid, excessive doses were
`given. Three feedings a week provided a dosage level equivalent to at least 50
`pzgm per day. This level was chosen with the thought that if the acid possesses
`the lowest activity yet reported for it——10 per cent as high as vitamin A5——we should
`still be providing about twice the vitamin A-activity considered to be adequate for
`the rat (2—2.5 pgm vitamin A per day).
`A Typical E’a:perz'ment.—The course of these experiments and the general nature
`of the results can best be introduced with such a typical experiment as shown in
`Figure 1. Two animals, litter mates, were placed on the deficient diet, supple-
`mented in one of them with vitamin A acid as described. Both animals grew at
`about the same rate for 5-6 weeks. Then the unsupplemented animal stopped
`growing, rapidly lost weight, and died on the 57th day of the diet. The other
`animal continued to grow regularly, and appeared to remain in prime condition,
`as the photograph taken on the 157th day of the diet is intended to show. On the
`same day this rat’s electroretinograms were recorded, as shown at the right of
`Figure 1, compared with those of a normal animal measured at the same time.
`This animal, though normal in weight and appearance, was highly night-blind.
`Its visual threshold——the luminance of a 1/so-second flash needed to excite a just
`measurable ERG—was 3.25 log units (about 1,800 times) above normal. As the
`figure shows, this animal yielded about the same ERG at log luminance 4 as the
`normal animal did at log luminance 0. Rhodopsin could be extracted from the
`retinas of animals in this condition in only 1-5 per cent of normal amounts. This
`degree of night-blindness was higher than we had ever achieved before in rats kept
`on the vitamin A-deficient diet without supplementation.
`On the following day these animals were sacrificed, and the retinal histology
`examined (Fig. 2).
`In the animal kept for 5 months on vitamin A acid, all the
`retinal tissues were normal in appearance, except for the visual cells. The pigment
`epithelium, bipolar layer (inner nuclei), and ganglion cell layer (not shown; see Fig.
`15) did not seem in any way altered.
`The nuclei of the Visual cells (outer nuclei) were considerably reduced in num-
`
`
`
`FIG. 1.—Nutritional activity of vitamin A acid. Litter mates had been placed on a
`vitamin A-deficient diet, and on the same diet supplemented with vitamin A acid. The
`rat given no supplement died on the 57th da of the diet;
`the animal receiving vitamin A
`acid continued to grow and remained in exce ent condition for the duration of the experi-
`ment, a little over 5 months. The picture of this animal was taken at the end of the
`experiment, as were the electroretinograms hown at the right, compared with those of a.
`normal animal. They show this rat to be highly night-blind:
`its visual threshold had
`risen 3.25 log units (about 1,800 times) above normal, and only just detectable ERG’s could
`be evoked at even the highest luminances.
`
`
`
`y of the night-blind animal shon in Fig. 1, compared with that of
`FIG, Z—R.etinal hlogl
`a normal animal. All the retinal tissues are normal except the visual cells, which are reduced
`in number and almost completely lack outer segments.
`
`
`
`590
`
`.
`
`BIOCHEMISTRYCDOWLING AND WALD
`
`PRoc.'N. A. s.
`
`ber, but otherwise -appeared normal. The reduction in the number of visual cells
`probably accounted also for a thinning out and compression of the layer of inner
`segments.
`A
`This retina “however lacked almost‘ completely the outer segments of the rods.
`Withithe rhodopsin, the organelles which contain and are largely composed
`of this
`hadalmost vanished.
`Suck 3i1~;aI.1jimiil?A-displays, as an isolatedf_'cond_i'tion freed of the complications of
`genet9.l.‘.tissue;§'¢lei3ifI‘y, the changes that characterizeutvhe development of dietary
`nightiz-ibliI_.p1_(j_lK1:1ess'A\_,:g;),,AI1 removal‘of the‘_fa1l«:in rhodopsin concentration, with
`associated
`then, deprived of the stabilizing effect of its
`prosthetic77gr’<)'u
`the decay?-rlf"epsin, with consefifient anatomical deterioration of
`
`
`
`flj/ctamzn A Alcohol
`f Supplement
`(4 animals)
`
`’A’
`
`60
`
`it
`
`1
`
`A
`
`l’
`
`A Days on Diet
`
`FIG. 3.‘—-Growth of animals on a vitamin A-deficient diet, compared with
`growth onvthe same diet supplemented with 50 ugm per day of vitamin A or
`vitamin A acid. The unsupplemented animals lost weight after 6 weeks on the
`diet, and all haddied by the end of the eighth week. The upplemented animals
`~ grew as well on vitamin A acid as on vitamin A.
`
`;
`‘”
`;
`
`the rod outer segments, in all likelihood the only tissue in the body of which opsin
`is an important component.
`‘
`In this instance the supplementation with vitamin A acid had resulted in an
`animal which appeared physiologically normal except for its night-blindness;
`biochemically normal, except for its lack of rhodopsin and opsin; and anatomically
`normal, except for the almost total loss of the outer segments of the rods. By the
`same token, this animal seemed to demonstrate that the only function in the body
`that requires vitamin A itself may be the formation of visual pigments. All the
`general tissue functions of vitamin A appear to be performed by vitamin A acid.
`These conclusions are examined in the remainder of this paper.
`Growth and Ma1Jntenance.—Weanling rats were divided into three groups, all
`placed simultaneously on the standard vitamin A-deficient test diet. One group
`
`H.‘,N:»._j58?
`
`
`1§
`
`x Vitamin A Acid
`supplement
`(to animals)
`
`'
`
`{Vitamin A
`Deficient
`(6anImo.1s)
`
`
` 0
`
`
`
`Von.-16, 1960
`
`BIOCHEMISTRY: DOWLING AND WALD
`
`591
`
`was supplemented with vitamin A, the second with vitamin A acid, both dissolved
`in cottonseed oil; and the third group was given the same amount of cottonseed oil
`alone.
`
`The growth of these animals is shown in Figure 3. For the first 5 weeks on the
`diet, all
`three groups grew about equally. Then the unsupplemented animals
`
`
`
`VitaminAAlcoholinliver(#9)
`
`8
`
`Rats fed
`Vitamin A
`Alcohol
`
`Vitamin A Acid.
`and
`
`/Vitamin A Del-‘tcie
`
`o
`
`20
`
`10
`Days on Diet
`FIG. 4.—Total vitamin A content of the liver in animals placed
`on a vitamin A—deficient diet, and on the same diet supplemented
`with 50 pgm per day of vitamin A or vitamin A acid.
`(These
`animals formed part of the same experiment as in Fig. 3.) The
`animals supplemented with vitamin A rapidly increased their
`liver stores. Those receiving vitamin A acid lost their initial
`stores of vitamin A as rapidly as those receiving no supplemen-
`tation.
`
`stopped growing, declined rapidly in weight, and by the end of the eighth week, all
`had died. The animals receiving supplements of vitamin A and of vitamin A acid
`grew equally well throughout the experiment (140 days). The small difference
`in average weight shown in Figure 3 (27 grams; 6.7 per cent) does not appear to be
`significant. Each of the supplemented groups spread considerably in weight, and
`
`
`
`592
`
`BIOCHEMISTRY: DOWLING AND WALD
`
`Pnoc. N. A. S.
`
`overlapped each other widely (vitamin A: 360-450 grams; vitamin A acid: 318-
`512 grams).
`In another experiment, after 135 days of a similar regime, the average
`weight of the animals on vitamin A acid was slightly greater than that of the animals
`receiving vitamin A (324 1314 grams).
`It may be concluded that the growth of the
`animals on vitamin A acid was entirely normal.
`Both groups of animals also appeared equally sound, externally and in the gross
`appearance of the internal organs on autopsy. The tracheal epithelium, examined
`microscopically, also appeared normal in both groups.
`To test further the biological eifectiveness of vitamin A acid, this supplement
`was withheld in other experiments until the 4th and 7th week of the vitamin
`
`50°
`
`§
`
`
`
`Vitamin A Acid. Rats
`
`Suppiements
`stopped
`
`resumed
`weight(grams) Us00
`
`
`me
`
`120
`
`140
`
`:60
`
`mo
`
`zoo
`
`220
`
`Days on Diet
`
`FIG. 5.—The effect of withdrawing the supplementation from animals
`growing normally on a vitamin A-deficient diet supplemented with ex-
`cess vitamin A acid. Within a few da
`these animals stopped grow-
`ing, and began to decline in weight.
`ithin 3-5 weeks, two animals
`had died displaying the signs of severe vitamin A deficiency. The
`third animal, on renewed supplementation with vitamin A acid,
`promptly recovered from its deficiency symptoms, and regained its
`normal weight.
`
`A-deficient diet, that is, until the animals had stopped gaining or were rapidly
`losing weight. On administration of vitamin A acid, such animals began at once to
`grow, and soon appeared identical with those supplemented from the start of the
`experiment. Also any signs of vitamin A deficiency that had developed on the
`deficient diet rapidly healed, except for occasional scarring of the cornea, owing to
`earlier xerophthalmia, which remained permanently.
`Vitamin A Storage and Depletz'¢m.—We had noticed that night-blindness develops
`equally rapidly in rats on a vitamin A—deficient diet, whether or not supplemented
`with vitamin A acid. During the first weeks on the diet such animals use up the
`supplies of vitamin A stored in the liver; and it seemed from this observation that
`
`
`
`VOL. 46, 1960
`
`BIOCHEMISTRY: DOWLING AND WALD
`
`593
`
`the rate of depletion of vitamin A in the liver is probably equally rapid, whether or
`not vitamin A acid is available.
`
`Figure 4 shows this to be the case. The vitamin A content of the liver was
`measured in animals divided into the three dietary groups already described;
`in-
`deed these animals formed part of the same experiment shown in Figure 3. Those
`receiving supplements of vitamin A (50 ugm per day) rapidly increased their stores
`
`0 Vitamin A- deficient
`0 V¢'£aminA acid
`I Vita/m'nA
`
`Days on Diet
`
`FIG. 6.—Visual thresholds of animals kept on a vitamin A-defi-
`cient diet, and on the same diet supplemented with vitamin A or
`vitamin A acid (same experiment as in Fig. 3). The threshold is the
`smallest luminance of a ‘/50 second flash of light needed to evoke a
`just-perceptible ERG. The vitamin A-supplemented rats formed
`the control group, whose thresholds are arbitrarily given the value 1
`(log threshold = 0). All other thresholds are expressed relative
`to these, and represent therefore increments of log threshold above
`the control level.
`In the animals supplemented with vitamin A acid,
`the threshold rises as soon and as rapidly as in those receiving no
`supplementation. All the latter group have died, however, by the
`end of the ninth week; whereas the vitamin A acid animals survive
`to grow more night-blind. The thresholds level off after 12-15
`weeks on the diet, at about 3.25 log units (about 1,800 times) above
`normal. At this time the retinas contain only 1-5 per cent of the
`normal amounts of rhodopsin.
`«
`
`in the liver. Those supplemented with vitamin A acid, however, lost their initial
`stores of liver vitamin A as rapidly as the rats receiving no supplements. Both
`groups also developed night-blindness at the same time. Vitamin A acid not only
`fails to contribute vitamin A to the liver, but seems to have no sparing action on
`the vitamin A already there. Since, however, the animals receiving the acid sup-
`plement are adequately maintained by it, except for their vision, the loss of vitamin
`A from the liver seems to be independent of demand. We shall have more to say
`of this relationship below.
`
`
`
`594
`
`BIOCHEMISTRY: DOWLING AND WALD
`
`Pnoc. N. A. S.
`
`The animals supplemented with vitamin A acid fail also to store this substance.
`Vitamin A acid is readily identified by the sharp absorption band at 573 mp which
`it yields in the antimony chloride test. We have never by this means been able to
`detect the acid in extracts of liver, kidney, or blood of animals receiving large
`amounts of this substance in the diet (cf. also Shannan’).
`’
`One consequence of the failure to store either vitamin A or the acid is that these
`animals, though normal in weight and in excellent health, respond almost im-
`mediately to interruption of the supplementation (Fig. 5). Animals growing well
`on vitamin A acid, on removing the supplement stop growing within a few days,
`
`0 Vitamin A-deficient diet 1’ 2
`
`Supplemented with w'taminA acid:
`o
`from start
`0 from 5th week
`e
`from 7th week
`
`,8
`E
`$
`0‘
`0
`\l
`
`~n
`
`O
`
`20
`
`40
`
`‘ 60
`
`80
`
`I00
`
`I20
`
`I40
`
`I60
`
`780
`
`Days on diet
`FIG. 7.—Development of night-blindness in animals on a vitamin A-deficient diet, and
`on the same diet supplemented with vitamin A acid from the start, or from the 5th week-
`when the animals had stopped growing—or from the 7th week, when they were declining
`rapidly in weight.
`In all these animals the visual threshold begins to rise at the same
`time, and rises equall
`rapidly. By the end of the 8th week, the unsupplemented animals
`have died; whereas t ose receiving vitamin A acid grow increasingly night-blind. After
`about 18 weeks on the diet the visual threshold of the latter group becomes constant at
`about 3.3 log units (about 2,000 times) above the normal level.
`
`decline rapidly in weight, and within 3-5 weeks die displaying all signs of severe
`vitamin A deficiency. That is, these animals respond as do young animals on a
`vitamin A-deficient diet after all internal stores of vitamin A have been exhausted.
`
`If the supplementation with vitamin A acid is renewed, even toward terminal
`A stages of the deficiency, such animals rapidly recover (Fig. 5).
`Night-Blindness.—We have already remarked that animals supplemented with
`vitamin A acid, though otherwise sound, become night—blind as rapidly as those
`receiving no supplementation. This isgshown in Figures 6 and 7.
`
`
`
`VOL. 46, 1960
`
`BIOCHEMISTRY: DOWLING AND WALD
`
`595
`
`Figure 6 shows the visual thresholds——the brightness of a ‘/50 second flash of light
`required to excite a just perceptible electroretinogram—of the animals Whose
`weights are shown in Figure 3. Those receiving supplements of vitamin A served
`as controls;
`their thresholds are arbitrarily given the value 1 (log threshold = 0),
`and all other thresholds are expressed as increments of log threshold above these
`control values.
`It is clear that the threshold rises at the same time and at about
`
`the same rate in the animals receiving no supplementation, and those receiving
`vitamin A acid; but whereas the former have all died by the 60th day of the diet,
`the latter survive to become increasingly night-blind.
`Figure 7 shows the results of a similar experiment, conducted with more animals
`and hence yielding more accurate averages. Again the unsupplemented animals
`and those given vitamin A acid from the start of the diet developed night-blindness
`at the same rate. Again the unsupplemented animals had all died by the 60th
`day; whereas the others, whether supplemented with vitamin A acid from the
`start, or from the 5th week—when they had stopped growing—or from the 7th
`week—when the weight was declining rapidly—continued to grow and to become
`more night-blind.
`In both these experiments the visual thresholds do not increase indefinitely,
`but level off after 12-15 weeks at 3.3—3.5 log units—about 2,000—3,000 times-
`above normal. Direct extraction of the retinas of such animals showed that they
`contain about 1—5 per cent of the normal amounts of rhodopsin.
`The electroretinograms of animals maintained on vitamin A acid are shown in
`Figmre 8. The stimulus was a 1/50 second flash of light, the luminance of which was
`varied in steps over a range of 7 logarithmic units (1 to 10 million). After 28
`days on the diet, the ERG is still normal. At 56 days, the animal is quite night-
`blind. The threshold has risen about 2.8 log units (about 500 times) ; at luminances
`above the threshold the response—both a- and b-waves—is greatly diminished;
`and a small inflection normally found on the downward sweep of the b-wave has
`separated off as a second positive wave. All these effects are characteristic of this
`stage of night-blindness in unsupplemented, vitamin A-deficient animals.”
`As the animals continue on vitamin A acid supplementation, the electroretino-
`grams undergo a second type of change. The threshold, as already described,
`becomes relatively constant by about the 120th day of the diet, at. about 3.2-3.5
`log units above normal.
`Increasing the intensity of the stimulus above the thresh-
`old level, however, now begins to have less and less effect. At 139 days, stimuli
`even 4 log units above threshold yield only a slow and diffuse b-wave .of low ampli-
`tude;
`indeed by that time the intensity of the stimulus hardly affects the form or
`height of the response (Fig. 8).
`It is as though, following upon the rise of threshold
`that marked the first stage of night-blindness, these animals now lose the capacity
`to generate an ERG. As Figure 8 shows, after 10 months on the regime, no response
`can be elicited at all.
`
`Figure 9 shows the relation between the visual threshold and the quantity of
`rhodopsin extracted from the retina. We had found earlier that in vitamin A-de-
`ficient animals receiving no supplementation, the logarithm of the threshold rises
`linearly as the rhodopsin content falls?
`In animals supplemented with vitamin
`A acid, this relationship can be extended further. As Figure 9 shows, it is main-
`tained over the whole range of the measurements.
`
`
`
`596
`
`BIOCHEMISTRY: DOWLING AND WALD
`
`P1100. N. A. S.
`
`Figure 9 involves a further comparison. The rhodopsin content of the retina
`falls in vitamin A deficiency;
`it falls also on exposure of the animals to bright light
`(light adaptation), rising again when the light is extinguished. That is, vitamin
`A deficiency and light adaptation are two ways of inducing night-blindness. How
`does the relationship between rhodopsin content and log threshold compare in
`both conditions? To answer this question a group of normal rats was highly light-
`
`FIG. 8.—Electroretinograms of animals on a vitamin Adeficient
`diet supplemented with vitamin A acid. Responses to a 1/50
`second flash of light, at luminances ranging over 7 log units (1 to
`10 million). The initial deflection downward is the negative a-
`wave;
`the later sweep upward is the positive b-wave. The first
`strip of records was made on the 28th day of the diet, when the
`ERG is still normal. On the 56th day the animal is typically
`night-blind:
`its threshold has risen about 500 times;
`the a- and
`; b-waves have decreased greatly in amplitude; and a small inflec-
`tion on the downward sweep of the b-wave has separated off as a
`second positive wave. By the 139th day, the threshold has risen
`only moderately further; but now increasing the brightness of
`stimulus above the threshold level has little effect. The retina
`is losing the capacity to generate an ERG. By the 288th day
`(10 months on the diet), no response can be elicited at the highest
`available brightness. The animal is now blind.
`
`adapted; then the light was shut off, and periodically an animal was anaesthetized
`and its ERG threshold determined. The same, and also other groups of animals
`were light-adapted similarly, and after the same intervals in darkness animals
`were killed, and their retinas extracted for rhodopsin.
`In this way the relationship
`between visual threshold and the retinal content of rhodopsin was measured, this
`time however in normal animals in the ordinary course of dark adaptation.” “
`
`
`
`VOL. 46, 1960
`
`BIOCHEMISTRY: DOWLING AND WALD
`
`597
`
`Figure 9 shows that a change of rhodopsin concentration in the retina, whether
`induced by visual adaptation or by vitamin A deficiency, has the same effect on the
`threshold.
`In both conditions the logarithm of the threshold rises linearly as the
`rhodopsin content falls.” The identity of these effects implies that up to this
`point in the development of vitamin A deficiency, the night-blindness is accounted
`for completely by the loss of rhodopsin from the retina.
`Anatomical Changes.—We have already seen that by the 24th week on vitamin
`A acid, the visual cells are considerably reduced in number, and most of the rods
`lack outer segments (Fig. 2). The disintegration of the rod outer segments seems
`
`100
`
`(DO
`
`
`
`0' Night Blindness
`
`0 Dark Adaptation
`
`8
`PercentHnodopsin 83
`
`
`o
`
`05
`
`1.0
`
`1.5
`
`2.0
`
`2.5
`
`' 5.0
`
`3.5
`
`L09 Threshold
`
`FIG. 9.-—Relat<ion between the rhodopsin content of the rat retina and the visual thresh-
`old, in animals night-blind owing to vitamin A deficiency, and in normal animals dark
`adapting after exposure to bright light.
`In both instances the same relationship is
`observed:
`the 10 threshold rises linearly (i. e., the log sensitivity-—log 1/threshold—falls
`linearly) with fal in rhodopsin concentration. This relationship is described by the
`equation log (It/I.,) = 3.6(R., — R.)/R.,,in which I. and E, are respectively the threshold
`and rho opsin concentration in dark adapted control animals, and I. and R: are respec-
`tively the thresholds and rhodopsir concentrations in vitamin A-deficient or partly dark-
`adapted animals.
`
`to accompany the loss of opsin. For example, in one experiment, in the 18th week
`of the deficiency (with vitamin A acid supplementation), when the threshold had
`risen over 3 log units, and the rhodopsin level was only a few per cent of normal, the
`opsin had declined to half the normal level.
`As the retina loses opsin, examination in the electron microscope shows progres-
`sive deterioration of the microstructure of the rod outer segments. The electron
`microscopy was performed in collaboration with Dr. I. R. Gibbons, and will be
`reported in detail elsewhere.” After removal of the cornea and lens, the Whole
`fundus of the eye was fixed for one hour in buflered osmium tetroxide, washed, de-
`
`
`
`598
`
`BIOCHEMISTRY: DUWLING’ AND WALD
`
`PROC. N. A. S.
`
`hydrated with acetone, and embedded in araldite resin. Ultrathin sections were
`cut on a Porter-Blum microtome, and examined in the R. C. A. EMU—3D electron
`microscope.
`Figure 10 shows a longitudinal section through portions of the rod outer seg-
`ments of a normal rat. As in-other animals, the outer segment consists of a stack
`of flattened sacs, wholly enclosed within an outer membrane. The outer segment
`is about 1.5 p wide and about 15 u long.
`It contains about 35 disks per g, or about
`525 disks in all.
`
`Figure 11 shows an early stage in the deterioration of this structure, typical of
`animals kept 3-4 months on the vitamin A-deficient diet supplemented with vitamin
`
`FIG. 10.—Electron micrograph of a longitudinal section through a rod outer segment, and por-
`tions of two neighboring outer segments, in the normal rat. As in other animals, the outer seg-
`ment consists of a stack of transverse, flattened sacs, entirely enclosed in an outer membrane.
`The rat’? o1(1)t6eb1' segment is about 1.5 u wide and 15 [1, long, and contains about 525 disks. Magnifi-
`cation 3,
`.
`
`A acid. Many of the retinal disks are still intact; but many others have segmented,
`or are in process of doing so, into groups of distended Vesicles and tubules.
`By 6 months on the regime, these changes have progressed further (Fig. 12).
`Few outer segments remain, and their internal structure is highly disorganized.
`They have lost also their characteristic cylindrical shape, becoming distorted and
`tending to collapse, often into roughly spherical shapes. By this time also the
`visual cells are considerably reduced in number, and their inner segments, formerly
`long and slender, have shortened and broadened (Fig. 2). The ultrastructure of
`the nuclei and inner segments of the visual cells, however, has not visibly changed.
`As animals continue longer on this regime, the visual cells undergo further altera-
`
`
`
`VOL." 46,1960
`
`BIOCHEMISTRY: DOWLIN6 AND WALD
`
`599
`
`tions. Figure 13 shows sections of- the retinas of two litter mates, .one'supplemented
`with vitamin-A, the other with; vitamin A acidffoi‘ 10‘months' on the otherwise.
`A-deficient diet. " The animal receiving vitamin'A acid no longer yielded an ERG‘
`in responseito the brightest‘ 1ights~~available iriour‘appa:raitu's (Fig. —8).
`In this
`retina,‘ the pigI'nent”epithe'lium’and the layers of=bi.polar and; ganglion cells still
`look normal. The visual cells, however, are reduced to ‘a single -row of nuclei, and
`the individual cells have contracted to a cuboidal. shape,‘ almost wholly occupied
`the nucleus,‘ with i no distinguishable inner or ou<ter1segment.> ‘‘ In this extraor-t
`dinary state the "pigment epithelium‘ lies inidirect contact with the“ ‘layer of visual
`cell nuclei.
`0
`
`
`
`
`FIG. 11.—Electron micrograph of a longitudinal section through the rod outer segment in a
`highly night-blind rat, typical of animals that had been 34 months on a vitamin A-deficiency
`diet supplemented with vitamin A acid. Some of the transverse disks appear inta.ct;- others
`have egmented into groups of distended vesicles and tubules. Magnification 113,000.
`
`Recovery.—On administration of vitamin A, animals which had become night-
`blind on the vitamin A—deficient diet supplemented with vitamin A acid recover
`in varying degree. When such animals have been on the diet up to 10 weeks, re-
`covery. ordinarily is -rapid and complete. Within several hours after feeding a
`large dose of vitamin A in cottonseed oil, the visual threshold——which may by then
`have risen about 2 log units above normal———begins to fall, and within 2—3 days has
`reached the normal level.
`~
`‘
`i
`'
`’
`
`After longer periods on the diet, recovery follows a more complex course. Figure
`14; shows the efiect of one intraperitoneal injection of 1 mgmof vitamin A on two
`animals that had been on the diet for: 25 weeks. These animals recover in two
`
`
`
`600
`
`BIOCHEMISTRY: DOWLING’ AND WALD
`
`Pnoc. N. A. S.
`
`the visual threshold first falls rapidly for 40-50 hours following the in-
`stages:
`jection, reaching a level about 1 log unit above normal;
`thereafter it continues to
`fall very slowly over a period of many Weeks of further vitamin A supplementation.
`In animals kept still longer on the diet, these two phases of recovery change in
`relative importance, the fast phase contributing less, and the slow phase accounting
`for more and more of the total change.
`It seems probable that these two stages of recovery are concerned with reversing
`the two phases we have described in the development of night-blindness: first, the
`rise of visual threshold associated with the decline of rhodopsin concentration owing
`
`FIG. l2.—Electron micrograph of the longitudinal section through a rod outer segment in an
`animal that had been 6 months on the vitamin A-deficient diet supplemented with vitamin A acid.
`Few rod outer segments remain in such a retina; and, as in this example, they are highly dis-
`torted in shape and microstructure. A few isolated groups of disks still appear, but mainly
`the internal structure has degenerated into distended tubules and vesicles. The outer segments
`have also lost their normal cylindrical shape, and collapsed to irregular ellipsoids or spheres.
`Magnification 76,000.
`
`to simple lack of vitamin A; later, the loss of opsin and anatomical deterioration of
`the rods.
`It seems reasonable to suppose that the fast phase of recovery involves
`the combination of whatever opsin remains with the vitamin A that is administered;
`whereas the synthesis of opsin and structural repair of the visual cells occupy the
`slow phase. Certain instances in which the recovery from experimental human
`night-blindness has exhibited similar fast and slow phases may perhaps be ex-
`plained in the same way.“
`In animals kept longer than 6 months on the diet, the visual threshold no longer
`recovers completely, even after months of vitamin A supplementation. This ir-
`
`
`
`VOL. 46, 1960
`
`BIOCHEMISTRY: DOWLING AND WALD
`
`601
`
`reversible aspect of night-blindness is probably associated with the actual loss of
`visual cells. Such cells, once lost, are never regained.
`A decline in the density of visual cells must raise the visual threshold, just as a
`decrease in the area of the visual field normally raises the threshold; and this
`should be a source of various degrees of permanent night—blindness.
`If indeed it is
`true that a decrease in the density of visual cells has about the same effect as a
`decrease in the area of the visual field, it should raise the threshold by a factor
`corresponding to the reciprocal of this decrease (Ricco’s Law) or the square root of
`the reciprocal (Piper’s Law). That is, in such a condition as shown in Figure 2,
`
`
`
`FIG. 13.—éections through the retinas of two animals—litter mates—both of which had been
`10 months on the vitamin A-deficient diet, supplemented with vitamin A (left), or with vitamin
`A acid (right). The retina on the left is normal; that on the right is also normal, except for its
`visual cells. The visual cells, onl
`about 1/10 of which remain, form a s