Biochem. J. (1965) 96, 475

Vitamin D3 and Calcium Absorption in the Chick

BY E. S. HOLDSWORTHDepartment of BiocheMistry, University of Adelaide, South Australia

(Received 29 September 1964)

1. An attempt has been made to locate the site of action of vitamin D3 as itaffects the translocation of calcium across the intestine. 2. Calcium appears to bepumped out of cells by a process dependent on energy from metabolism. 3. Theeffects of cold, inhibitors and vitamin D3 on the translocation of calcium by evertedsacs ofintestine were studied and compared with results obtained in vivo. 4. Amodelwas proposed to explain the results which suggests that vitamin D3 inhibits a

metabolically operated pump that returns calcium from the mucosal cell to thelumen. 5. Some observations on the effect of sodium lauryl sulphate on the trans-location of calcium in vivo and in vitro are reported.

It has been firmly established that vitamin Dincreases the transport of calcium from the smallintestine of many species (Harris & Innes, 1931;Nicolaysen, 1937; Greenberg, 1945; Albright &Reifenstein, 1948; Harrison & Harrison, 1951;Migicovsky, 1957; Wasserman, Comar, Schooley& Lengeman, 1957; Schachter & Rosen, 1959;Coates & Holdsworth, 1961; Sallis & Holdsworth,1962). From experiments with rat intestine invitro Harrison & Harrison (1960) suggested thatvitamin D increases the rate of diffusion of calciumthrough the intestinal mucosa, possibly by thealteration of the permeability of intestinal cells tocalcium. A similar conclusion was arrived at byWasserman & Kallfelz (1962), who also demonstra-ted that the usual criteria of active transport asdeveloped by Ussing (1949) were unable to accountfor the facts of calcium transport. A model systemfor transcellular transport that has been developedby Parsons (1963) appears to be of general applica-bility, and the present paper uses this model in anattempt to explain the mode ofaction ofvitamin D3.During an investigation into the role of bile and

detergents on calcium transport, it was noticed thatsodium lauryl sulphate improves calcium uptake inrachitic chicks (Webling & Holdsworth, 1965). Theeffect of this substance on calcium transport hasbeen studied since it may provide a clue as to howvitamin D3 achieves its effect.

EXPERIMENTAL

Birds and diet. White Leghorn cockerels were raised on adiet deficient only in vitamin D for a period of 4 weeks, asdescribed by Sallis & Holdsworth (1962). To study theeffect of vitamin D, the rachitic chicks were dosed with200i.u. of vitamin D3 in arachis oil by stomach tube 16hr.before the experiments.

Perfusions in vivo. Through the distal third segment ofthe small intestine of the chick was circulated 10ml. of0-15m-NaCl containing glucose (0.02M) and CaCl2 (either1mM or 10mM; 10,uc of 45Ca) by a peristaltic pump (Sallis& Holdsworth, 1962). Sodium lauryl sulphate, when addedto the perfusion fluid, was used at a concentration of 8mm.Samples of blood and circulating fluid were taken atintervals. At the end of the experiment the segment of gutwas placed in ice-cold 0 15M-NaCl and then washed throughwith a stream of ice-cold 0-15M-NaCl. The segment wasthen cut open and pressed between soft tissue, and themucosal cells and muscle were separated. The absorptionof calcium was determined by measuring the 45Ca in thecarcass.

Transport of calcium in vitro. Segments of the distalthird of the small intestine of the chick were everted andincubated as described by Sallis & Holdsworth (1962).For measurement of transport at 00, the flasks containingthe everted sacs were surrounded by ice. In some experi-ments the gut was not everted and on these occasions theserosal (outside) surface of the sac was immersed in the5ml. of Krebs-Henseleit bicarbonate buffer (Krebs &Henseleit, 1932) containing glucose (0-02ar) and 45CaC12(2mm), and the sac (mucosal surface) contained lml. of thesame buffer containing non-radioactive CaCl2 (2-25mm).The flasks containing the sacs were shaken and gassed with02+ C02 (95: 5).At the end of the incubation period, the outside of the

gut was washed with a stream of ice-cold 0 15M-NaCl anddried with tissue. The inside fluid was removed, the sacwas rinsed with lml. of 0 15m-NaCl and the combinedliquids were examined for 45Ca content. The segment waseither taken as such or split longitudinally, the mucosalcells were scraped off and the 45Ca contents of the cells andmuscle were measured.To preload the everted sacs with 45Ca they were incubated

in the usual way either at 00 or at 380 for 30min. Themucosal surface was washed with ice-cold Ol15m-NaCl, andthe serosal fluid was removed, rinsed out with lml. of015M-NaCl and replaced with lml. of Krebs-Henseleitbuffer containing glucose (0-02rM) and non-radioactiveCaCl2 (2mmr). The sac was placed in 5ml. of this same buffer

475

E. S. HOLDSWORTHand reincubated at O or at 380 for 15min. The serosalfluid, mucosal cells and muscle were treated as describedabove.

In all these experiments in vitro a reproducible length ofgut was obtained (10-12cm. long) by using chicks at thesame stage ofdevelopment and taking a segment ofintestinefrom the point of attachment of the yolk-sac down to thecaecum. The dry weight of cells and muscle from suchsegments differed little (see Table 3), and it was found thatit made no difference to the results to correct for this slightvariation in weight. For this reason all results are expressedas for a segment of intestine.

Incorporation into mucosal cells. The mucosal cells ofthe lower third of the small intestine were scraped off withthe round end of a spatula and placed in ice-cold Krebs-Henseleit bicarbonate buffer containing glucose (0-02M).The cells were washed once with the buffer and resuspendedin the same medium to give a concentration of 50mg. ofprotein/ml. To 2-5ml. (in duplicate) of such a suspensionwas added 45CaC12 (1mM; 0-2,uc) and the mixture incubatedat 380 or kept on ice for 15min. The cells were removed by ashort high-speed centrifugation for approx. 2 min. at30000g and the 45Ca in the supernatant was determined.After being washed twice with ice-cold buffer, cells that hadtaken up 45Ca at 0° were incubated in Krebs-Henseleitbicarbonate buffer containing glucose (0-02M) but nocalcium. The 45Ca in the supernatant and cells was deter-mined.

Incorporation into mitochondria. Mucosal cells preparedas above were homogenized in 0-25M-sucrose with a Potter-Elvehjem homogenizer and the fraction sedimentingbetween 1000 and 100OOg was obtained. The pellet wasresuspended in 0-25M-sucrose containing 10,umoles of trisand 10,amoles of succinatefml. (final pH7-2). The con-centration of protein was adjusted to 2-5mg./ml. and thatof 45CaC12 to lmM (0-2,uc). After incubation for 15min. at380 in 02+ CO2 (95:5), the mitochondria were centrifugeddown and washed once with ice-cold 0-25M-sucrose and the45Ca was determined.

Preparation of cell fractions. Mucosal cells were scrapedoff the segments of intestine with the round end of a spatula.In most experiments 45Ca was determined in the mucosal

cells and the residual muscle. In some experiments themucosal cells were homogenized with 0-25M-sucrose andfractions prepared by centrifugation at 600g for 10min.,2000g for 10min., 20000g for 20min. and 100000g for60min. The pellet sedimenting between 600g and 2000gwas resuspended in 20% (w/v) sucrose and centrifuged at1000g for 10min. and then at 3000g for 10min. The 45Cawas determined in all these fractions.

Determination of 45Ca. (a) Blood. A 0-2ml. sample ofplasma was treated with 0-2ml. of 10% (v/v) HC104 andcentrifuged, and 0-2ml. of the supernatant was added to5ml. of scintillator fluid (Bray, 1960).

(b) Serosal fluid. A 01-ml. sample was added directly tothe scintillator fluid.

(c) Intestinal samples. Organic matter from intestinasegments, mucosal cells, muscle and mitochondria wasdestroyed by wet oxidation with conc. HNO3 and HC104.After evaporation to dryness, the residue was dissolved inwater and a portion added to the scintillator fluid.

(d) Carcasses. To determine the 45Ca taken up duringperfusion, the carcass, after removal of the perfused loop,was ashed in a silica dish, converted into chlorides byevaporation with cone. HCI and dissolved in 100ml. ofwater. The radioactivity in 0-5ml. portions was determinedby scintillation counting.

(e) Scintillation counting. All samples were counted inthe scintillation fluid developed by Bray (1960) by using anEkco N664A counter. An internal standard of 45Ca wasadded after counting the sample and suitable correctionswere made for quenching.

RESULTS

Perfusions in vivo. It was found that the onlyreliable method ofmeasuring the amount of calciumabsorbed from the perfusion fluid was to determine45Ca in the carcass. Estimation ofcalciumuptake byestimating 45Ca in the circulating fluid presupposesan accurate knowledge of the volume change of thisfluid, which could not be determined with accuracy.The results obtained at two concentrations of

Table 1. Absorption of calcium by small intestine during perfusion in vivo

Distal segments of small intestine from rachitic or vitamin D3-treated chicks were perfused with 10ml. of 0-15M-NaCl containing CaCl2 (1-0 or 10mM; 10,eC of 45Ca) and glucose (0-02M). Perfusion was for 30min. at 380 andblood samples were taken at 10 and 20min. The results are mean values + S.E.M. for six chicks.

Concn.of

CaCl2(mM)

1-0Treatment

RachiticVitamin D3

10 RachiticVitamin D3Rachitic+sodium laurylsulphate (8mM)

45Ca incarcass

(,ug./30min.)3-1+1-738+9

22+ 8190+ 43290+ 67

45Ca in mucosa([kg./30min.)

Cells Muscle2-8+1-1 2-0+0-835+1-5 19+0-617+3-921 + 5-148+ 92

11+2-515+4-116+4-5

45Ca in plasma(counts/min./ml.)

lOmin. 20min.8300* 770038900 55900

41001690029000

52001870026300

* The 45Ca absorbed at mm-CaCl2 had 10 times the specific activity of that at lOmm-CaCl2.

1965476

VITAMIN D3 AND CALCIUM ABSORPTIONCa2+ are shown in Table 1. Vitamin D3 greatlyincreased the translocation of calcium into thechick but there was no significant increase in theamount of calcium bound by the mucosal cellsunder the steady-state condition reached after30min. perfusion. Sodium lauryl sulphate alsoincreased the translocation ofcalcium in the rachiticchick and in this instance there was a markedincrease in the calcium taken up by the mucosalcells (Table 1).

Studie8 in vitro. (a) Everted sacs were kept at00 and the amount of 45Ca transferred to the serosalfluid and taken up by the mucosa was measured.Vitamin D3 had no effect on these processes (Table2). When the everted sacs were preloaded with45Ca in this way and then rinsed and placed inmedium containing non-radioactive Ca2+, theexchange or transport of 45Ca could be measured inthe mucosal and serosal fluids. Table 2 shows that,for mucosa preloaded at 0° and reincubated at 00,

the process of exchange with the mucosal fluid or

transport into the serosal fluid was the same forgut from both rachitic or vitamin D3-treatedchicks. Table 2 also shows that cells preloaded at 00

and reincubated at 380 transported or exchangedmore 45Ca at the higher temperature, but againthere is no observable effect of vitamin D3.

(b) Everted sacs incubated at 380 show thatvitamin D3 increases the amount of 45Ca transferredto the serosal fluid and also increases the 45Ca foundin the mucosal cells and muscle. These results(Table 3) are in agreeement with our previous

findings and those of other workers (Sallis &Holdsworth, 1962; Harrison & Harrison, 1960;Schachter & Rosen, 1959). WVhen everted sacs were

preloaded with 45Ca at 380 and then reincubatedin medium containing non-radioactive Ca2+ at 00,

the 45Ca was transferred to the serosal and mucosalfluids in amounts proportional to the initial 45Cacontent of the sac. This last value was obtained bysummuing the 45Ca contents of mucosal and serosalfluids and mucosal cells and muscle. Thus theratio of 45Ca obtained by dividing the uptake ofthe sacs from vitamin D3-treated chicks by theuptake of the sacs from rachitic chicks was 3 1,and the ratio of transfer to serosal fluid by sacs

from vitamin D3-treated and rachitic chicks was

5-5 during the loading period and 3-9 on reincuba-tion. Thus the vitamin D3 effect was obtainedduring reincubation at 00 but appears to be relatedto the initial uptake of 45Ca under the influence ofvitamin D3 at 380 (Table 4).Everted sacs preloaded at 380 and reincubated

at 380 showed greater exchange and transfer toserosal fluid than those reincubated at 00, butagain the greater transfer to serosal fluid observedwith sacs from vitamin D3-treated chicks appears tobe due to the greater uptake during the preloading

o 0

CO S

.016

,, O-.:

0

000

t .5 E

COC)

0~.) SO

o0

02) O~S

0

O Oa o

3 3S.+

C).5 cOv

,

-5V

(D

Ca

0

0

00

CO

.5

-z

0

c¢3

0

+0o

oC)

4-DA

CO

*SO

*S

90 ,t4

-4

CPS

04 ;.40 C

m 84bo o

Q Qr

fi

umPs

C44.40)

0)5

OP0)c3k

Q CN 8441 =

Swp-I--,

0L) 0.

0

.-45

01)0

P_1

e$ 4

S

0

0 boc) -

ceu:Ul

O m

O

oq co0

>0

5+1 +1

H- CO O .¢

c3c00

+I +I

Co CO

Ps

-oo-

c3 z

aE-00+1+1"

4a3

a 0

+It+l +- Om-

Ca

Sog_H"

'041 0z.

Vol. 96 477

E. S. HOLDSWORTHTable 3. Uptake and transport of calcium at 380 by everted sacs of small intestine

Everted sacs of small intestine from rachitic or vitamin D3-treated chicks were incubated for 30min. at 380 inKrebs-Henseleit bicarbonate medium containing 45CaCl2 (2mM) and glucose (0.02M). The 45Ca uptake by thecells and muscle of the mucosa and of the serosal fluid was determined. Accumulated results (means + s.E.M.) ofseveral experiments (32 sacs) are given. Only six sacs were used in the experiments with added sodium laurylsulphate.

RachiticVitamin D3-treatedRachitic+ sodiumlauryl sulphate (8mM)Vitamin D3-treated+sodium lauryl sul-phate (8mM)

Mucosal cells Muscle layer

45Ca accumulated(,ug./30min.)2-5+0956-9+ 17

- 10-4+11918-4+2-9

Drywt.(mg./mg.)253+38241+43

45Ca accumulated(,ug./30min.)31+1--1114+ 2*1

- 5-4+1-811*5+2-0

Serosal fluid45Ca accumulated

(p(g./30min.)1.9+0*598-2+2*1

2-2+0*403-9+0*58

period by the sacs from vitamin D3-treated chicks.This can be seen from the value for the transfer of45Ca as a percentage of the 45Ca content after thepreloading period and in Table 4 referred to as

transfer as a percentage of total uptake.(c) The transport of 45Ca and the uptake of 45Ca

by everted sacs was studied in the presence andabsence of 2mm-iodoacetate (Table 5). Iodoacetatehad only a slight effect on the intracellular 45Caof the mucosal cells of sacs from vitamin D3-treatedchicks but it did prevent the further translocationof this 45Ca to the serosal fluid. With everted sacs

from rachitic chicks, iodoacetate caused an

accumulation of 45Ca in the mucosal cells, and againfurther transport of the 45Ca was blocked. Theamount of 45Ca found in the muscle layer of sacs

from rachitic chicks was increased by iodoacetatewhereas iodoacetate caused a decrease in theamount of 45Ca found in the muscle layer of thesacs from vitamin D3-treated chicks.

(d) When sodium lauryl sulphate was added to themucosal fluid there was a great increase in theamount of calcium taken up by the mucosal cells.Only a small proportion of this calcium was trans-ported to the serosal fluid even when segmentsfrom vitamin D3-treated chicks were studied(Table 3).

(e) The effect of increasing the concentration ofmucosal and serosal Ca2+ on the translocation of45Ca was studied. With increasing Ca2+ concen-

tration the ratio of transfer to serosal fluid by sacs

from vitamin D3-treated and rachitic chickssteadily decreased (Table 6).

(f ) Translocation of 45Ca in the opposite directionto that usually studied was investigated by havingsacs of intestine treated in usual manner exceptthey were not everted. Transfer from serosal tomucosal surface is shown in Table 7, together withresults obtained with the same batch of chicks but

by using everted sacs. As expected, there is lesstranslocation of 45Ca from serosal to mucosalfluids than in the opposite sense. The most impor-tant observation is that transport with non-

everted sacs was not affected by vitamin D3treatment.

Distribution of 45Ca in the small intestine.Table 8 shows the 45Ca content of broken mucosalcells and muscle after preloading the segmenteither in vivo or in vitro. The greatest concentrationof 45Ca in mucosal cells always occurred in a

fraction that sedimented between 1000 and 3000gin 20% sucrose. The large increase of calciumtaken up by mucosal cells in the presence of sodiumlauryl sulphate is evident (Table 8), and again thefraction sedimenting between 1000 and 3000gcontained most of the radioactivity. Attempts toidentify this fraction have not been successful.Electron-microscope pictures (kindly produced byDr G. Rogers, University of Adelaide) showed thepresence of structures resembling swollen and dis-torted brush borders, but the preparation alsocontained mitochondria and cell-membrane frag-ments. When intestine was preloaded with 45Ca,it was found that mucosal cells and muscle fromeverted sacs always contained more calcium thantissue loaded by circulating 45Ca in vivo even thoughthe same solution of 45Ca was used for the twopreparations.

Uptake of 45Ca by mucosal cells. Cells scrapedfrom the small intestine of rachitic or vitaminD3-treated chicks were incubated with 45Ca eitherat 00 or at 380. The uptake of 45Ca by the cells isshown in Table 9. Mucosal cells took up much more45Ca at 00 and there was little difference betweenrachitic and vitamin D3-treated cells. When cellsloaded at 0° were reincubated in calcium-freemedium at 380 a considerable amount of 45Ca wasreleased. In a total of eight experiments where

478 1965

Drywt.(mg./mg.)113+ 13120+ 17

VITAMIN D3 AND CALCIUM ABSORPTIONmucosal cells were incubated at 380 with 45Ca, six

?4 - experiments showed slightly more 45Ca accumulated.S.> $ in the cells from vitamin D3-treated chicks than inCa Ca Co 0 =2 m csethose from rachitic chicks (ratio 1.3-1-5) and in

two experiments there was no difference in uptake0fOo0 g between the two kinds of cells.

5o 0 C Results obtained with mitochondria were very

A 0 0 a variable: out of 13 experiments, eight showed ati .lgreater uptake by those from vitamin D3-treated0 Ca chicks than by those from rachitic chicks (ratio

onQQQ 1-31-7) and five showed no difference. No dif-Ca -P *=~z - ferences were found in the uptake of 45Ca by brush

vSo borders prepared by the method of Miller & Crane& o +1+

', ,"

..5

aQ ' m o (1961) from rachitic or vitamin D3-treated chicks.

DISCUSSION

0t S.g t cz OTo account for the experimental observations++1+ and to locate the site of action of vitamin D3,

o .S <? Cl.° various models for transcellular transport, basedc L on the ideas put forward by Parsons (1963), were

¢ D o & W O considered. Essentially a sheet of mucosal cells.% .S o was considered to have two faces, one bordering

e.:0 = i ; <:)mthe lumen of the intestine and the other, serosalCB ssurface, facing the muscle of the gut. Each face

0 4) O OHXUwasassumedtopossesstwoindependentproperties:

0 -o Q e i 8 l a t (1) a substrate pump operated by energy obtained2 a . O 0 CZ from metabolism; (2) a diffusion leak working in the

.sS p S 8o > gopposite sense to the pump. When consideringCB E e < o A Sthe active transport of sugars and amino acids,

o > Q ° Q Parsons (1963) considered the pumps to be directedX , .5s . ~ uC P

z intracellularly; however, when considering a

Ca-Eea R o o #substrate such as Ca2 , which is usually presentS -. +18+ only in low concentrations inside cells, there is noE 0 reason why the pumps should not be considered as

S ' ~ '8 removingC2+ from the cell. Vitamin D3 could bezi - CBv~S considered to enhance or inhibit either or both ofczQ. ^ ) . taSO O the pumps, or either, or both, ofthe two leaks. Thus

.t .5 M e+1 +1 a large number of model systems were considered,Ev .D 8$D but the experimental results obtained allow the

0 ,5 S - elimination of many of these alternatives, and the45'4 x _action of vitamin D3 can best be accounted for

by considering it to inhibit the metabolicallyg +1 C: _ ae 2 e °driven pump P1 (Scheme 1). The evidence to sup-

o _ 0 o port this hypothesis is as follows.0 0

8 w a e S ~ ^ s (i) If pump P1 were inhibited (cf. Scheme 1)s}i °~ e cc 8 there would be a rise in intracellular calcium

P4 t- 00g S S e e > concentration and this would reach the serosali, E + + surface via pump P2 and diffusion leak K2. The

,s g o C m lo overall result would be an increased translocation of0Woi > . > e ¢ calcium such as that found with everted sacs from

EQ . vitamin D3-treated chicks (Table 3).CB X,e(ii) If pump P1 were enhanced by vitamin D3

E--iE there would be a greater return of calcium to theUpa> lumen and hence a decreased translocation of)S. calcium, which is contrary to the experimental

CB (iIuweibdfindings.g V. ~~(iii) If PUMP P2 were inhibited there would be a

Vol. 96 479

Table 5. Effects of iodoacetate on calcium uptake and transport by everted sacs of small intestine

Everted sacs of small intestine from rachitic or vitamin D3-treated chicks were incubated for 30min. at 380 inKrebs-Henseleit bicarbonate medium containing 45CaC12 (2mm) and glucose (0-02M). Some flasks containediodoacetate (2mM). The results are mean values + S.E.M. for eight sacs.

45Ca translocated (,ug./30min.)

No inhibitor

Serosal fluidMucosal cellsMucosal muscle

Rachitic2*46+ 0492*62+ 0-613-54+ 0.72

VitaminD3-treated1168+ 1976*88+ 10211-74+ 2*12

2mM-Iodoacetate present

Rachitic2*06+ 0385-58+ 0-895-66+0.94

VitaminD3-treated3.26+0 547-22+ 1-276*36+ 1-15

Table 6. Influence of Ca2+ concentration on calciumtransport by everted sacs of small intestine

Everted sacs of small intestine from rachitic or vitaminD3-treated chicks were incubated for 30min. at 380 in015M-NaCl containing glucose (0-02M) and the statedconcentration of Ca2+ (with 4,o of 45Ca). The results aremean values+ S.E.m. for four sacs.

45Ca in serosal fluid(,tg.)

VitaminRachitic D3-treated16-9+2*4 25-4+4_46-5+1-3 15-7+2-22-1+0-29 7-9+1-9

0.51+0-27 4-2+2.9

Ca transported bysacs from vitamin

D3-treated chicks/Catransported by sacsfrom rachitic chicks

ratio (range)1-2-1-82-5-3-03-7-5-64-2-9-8

* At this concentration some of the mucosal cells were

shed into the mucosal fluid.

Table 7. Transport of calcium by non-everted sacs

of small intestine

Sacs of small intestine from rachitic or vitamin D3-treated chicks were either everted or non-everted andincubated under the usual conditions (see Table 2). Themean (±S.E.M.) serosal transport of 45Ca (of six sacs) was

determined.Serosal transport of 45Ca

(uig./30min.)

RachiticVitamin D3-treated

Everted sacs

2*15+0-398-20+ 0*95

Non-everted sacs

1-32+ 0-611-43+ 0 49

rise in intracellular calcium concentration butthere would be a decrease in the amount of calciumtranslocated. Since vitamin D3 increases calciumtransport it cannot inhibit pump P2.

(iv) If pump P2 were enhanced there would be adecrease in intracellular calcium concentrationwith an increase in calcium translocation. Witheverted sacs, vitamin D3 always caused an increasein intracellular 45Ca concentration (Table 3),and therefore it appears that vitamin D3 does notenhance pump P2. Further, when everted sacswere preloaded with 45Ca at 00 and then incubatedat 38° there should have been a noticeable increasein translocation of 45Ca ifvitamin D3 was enhancingpump P2. No such increase could be detected(Table 2).

(v) If vitamin D3 increased the diffusion leakK1 (or K2 or both) this would lead to increasedtranslocation ofcalcium. It would be expected thatthis increased permeability due to vitamin D3would also be evident at 00. The experimentalresults show no difference in 45Ca transport betweeneverted sacs from rachitic and vitamin D3-treatedchicks at 00 (Tables 2 and 4). No effect of vitaminD3 was found on the movement of 45Ca from theserosal to the mucosal compartments (Table 7).This again suggests that the vitamin does notalter diffusion leaks K2 and K1, thus increasing thepermeability of the cell to calcium.

(vi) Ifpumps P1 and P2 were both inhibited therewould be an increase in intracellular calcium con-centration but transport across the cell would belimited and dependent solely on diffusion viadiffusion leaks K1 and K2.

Cooling to 0° should inhibit the metabolicallydriven pumps, and under these conditions evertedsacs from rachitic and vitamin D3-treated chickswere found to take up and transport the sameamounts of45Ca (Table 2). It appears to be necessaryto have energy produced from metabolism to beable to show an effect of vitamin D3.

Isolated mucosal cells took up 45Ca at 00 andagain no effect of vitamin D3 was apparent. Whenthese cells that had taken up 45Ca at 00 werereincubated at 380, the 45Ca was lost from the cells(Table 9). Further, isolated cells took up more

Conen.of

CaCl2(mM)100*101.00.1

480 E. S. HOLDSWORTH 1965

VITAMIN D3 AND CALCIUM ABSORPTION 481Table 8. Distribution of calcium in the mucosa of the small intestine

Three segments of small intestine from rachitic or vitamin D3-treated chicks were loaded with 45Ca either byperfusion for 30min. (as in Table 1) or by incubation of the everted sacs (as in Table 2). The medium used was0-15M-NaCl containing glucose (0-02M) and CaCl2 (10mM; 10,uc of 45C in lOml.). The segments were rinsed withice-cold 0*15M-NaCl and the cells and muscle separated. The cells were homogenized with 0-25M-sucrose and thehomogenate was separated by differential centrifugation. Cells and muscle for three segments were combined(to produce sufficient material) but the results are expressed as for one segment.

45Ca content (,ug. accumulated/30 min.)

Perfused intestine Everted sacs

FractionDebris (600g for lOmin.)Centrifuged at 2000g for lOmin. (refrac-tionated in 20% sucrose to give fraction1000-3000g for lOmin.)

'Mitochondria' (20000g for 20min.)'Microsomes' (104000g for 60min.)Supernatant (104000g for 60min.)MuscleTotal Ca

Vitamin Sodium laurylRachitic D3-treated sulphate

1-68*9

2-00-80*9

13_828-0

0*97.5

2-00-80-6182300

8-228*1

2*71-41-0

17*358-7

VitaminRachitic D3-treated

2-36-9

2-00-80-8

21*834-6

4-111-3

3.30-70-9

31-752-0

Table 9. Uptake and release of calcium by i8olatedmucosal cells of the small intestine

Mucosal cells were scraped from the distal third of thesmall intestine of rachitic or vitamin D3-treated chicks.Suspensions of the cells in Krebs-Henseleit bicarbonatemedium containing glucose (0-02M) contained 50mg. ofprotein/ml. The cells in 2-5ml. were made 10mM withrespect to 45CaCl2 and incubated for 15min. at 00 or 38°.Uptake of 45Ca was obtained by counting the residual 45Cain the supernatant after centrifugation. Cells were washedtwice with ice-cold buffer. The cells that had accumulated45Ca at 00 were then incubated in the above medium forlOmin. at 38°. Results are the means of duplicate deter-minations.

45Ca content(,g./125mg. of protein)

Uptake by cellsFirst washingSecond washingRelease on reincuba-tion at 380

VitaminRachitic D3-treated

00 380 00 38080-1 40-0 90-7 41-51-2 23-5 1-3 26-40-1 10-0 0-3 9*4

54-6 61-0

45Ca at 00 than at 380 and this finding supports theidea that calcium is pumped out of the cells bymetabolically activated pumps. Similar observa-tions were made with rat mucosal cells by Rasmus-sen, Waldorf, Dziewiatkowski & De Luca (1963).

16

Lumen Serosa

Scheme 1. Translocation of calcium by the mucosal cells ofthe small intestine. K1 and K2 are diffusion leaks, and

P1 and P2 are pumps operated by energy from metabolism.Cai, Intracellular calcium.

Sallis & Holdsworth (1962) showed that theenergy for calcium transport can come from glycoly-sis and that iodoacetate would inhibit this process.It seems likely that iodoacetate would inhibitpumps P1 and P2. lodoacetate caused the intra-cellular 45Ca concentration of everted sacs fromrachitic chicks to rise, but prevented the transportof this 45Ca to the serosal fluid (Table 5). This wasthe finding predicted above. When sacs fromvitamin D3-treated chicks were studied (wherepump P1 was considered to be already inhibited byvitamin D3) the further inhibition by iodoacetatedid not lead to a significant increase in intracellular45Ca concentration but, by inhibiting pump P2,did prevent this 45Ca reaching the serosal fluid.This difference in response of mucosal cells fromrachitic and vitamin D3-treated chicks to iodo-acetate is further evidence for locating the site ofaction of vitamin D3 at pump P1. If vitamin D3produced its effect by increasing diffusion leak

Bioch. 1965, 96

Vol. 96

E. S. HOLDSWORTH

K1, then inhibition of pumps P1 and P2 by iodo-acetate would be expected to lead to a rise in intra-cellular 45Ca concentration in sacs from bothrachitic and vitamin D3-treated chicks, but thiswas not observed (Table 5).

(vii) Increasing the concentration of calcium inthe mucosal fluid would lead to a greater diffusionof calcium into the cell via diffusion leak K1 andsubsequently to the serosal fluid via diffusion leakK2 and pump P2. At high concentration of calciumthe capacity of pump P1 might be overwhelmedand in these circumstances the difference betweenthe amounts of calcium translocated by gut fromrachitic and vitamin D3-treated chicks would beless marked. That this is so is shown in Table 6,and this is in agreement with similar findings bySchachter & Rosen (1959), who worked with ratintestine. There is no hint even at 0 1M thattranslocation of calcium is reaching saturation, andthis is consistent with entry by diffusion, whereasdiffusion facilitated by 'carrier' molecules would beexpected to become saturated.The above arguments show that the hypothesis

that vitaniin D3 inhibits pump P1 can account forthe experimental observations reported in thepresent paper. Webling & Holdsworth (1965)have shown that bile and sodium lauryl sulphatewill increase calcium translocation in a mannerthat resembles that of vitamin D3. Observations inthe present paper show that sodium lauryl sulphatediffers in some respects from vitamin D3 in itseffect on calcium transport. Table 1 shows thatthis detergent increases 45Ca transport in vivo butat the same time causes a greater accumulation of45Ca in the mucosal cells than does vitamin D3(Table 1). A more marked accumulation of 45Cain mucosal cells was seen when sodium laurylsulphate was used with everted sacs (Table 3). Itcould be considered that sodium lauryl sulphatewas acting as an inhibitor of pumps P1 and P2 in asimilar manner to iodoacetate. However, theintracellular 45Ca obtained with sodium laurylsulphate was much greater than that obtained withiodoacetate. Also, it is difficult to see why sodiumlauryl sulphate is so effective in the translocation of45Ca in vivo if it is inhibiting pump P2, althoughexperiments with everted sacs did show that less45Ca appeared in the serosal fluid of sacs fromvitamin D3-treated chicks when sodium laurylsulphate was present (Table 3). An alternativesuggestion (Webling & Holdsworth, 1965) is thatsodium lauryl sulphate (or bile) converts thecalcium into a lipid-soluble complex that is morereadily taken up by the lipophilic cell membranes.For significant translocation of calcium from lumento serosal fluid to take place the calcium in themembranes must be released to the serosal fluid.With the small volume of fluid within everted sacs

conditions would be unfavourable for calciumtranslocation (Table 3), whereas in vivo, whereplasma is continually flowing past the cells, thereis a large volume into which the calcium can bedistributed and this could account for the markedtranslocation of calcium brought about by sodiumlauryl sulphate in vivo (Table 1).A theory to explain the increased transport of

calcium brought about by sugars such as lactose orsucrose has been put forward by Wasserman (1964).He has correlated the effect of these substances ontransport with their degree ofinhibition ofoxidativephosphorylation. No detailed model for transportacross the cell was proposed, but the suggestionthat lactose is inhibiting the energy supply iscompatible with the model proposed in the presentpaper.The model in Scheme 1 shows that the diffusion

leak and the metabolically operated pump arelocated at different sites on the cell membrane.However, a leak and a pump can be visualized asoccurring at the same site, i.e. diffusion could takeplace via the pump if this were not completelyefficient. Thus substances that alter calciumtransport would produce their effects by increasingor decreasing the efficiency of pumps P1 and P2.The dangers of using the system in vitro as a

model for what happens during calcium transportin vivo were pointed out by Sallis & Holdsworth(1962). However, vitamin D3 does appear to haveeffects when using the isolated everted intestinequalitatively similar to those obtained in vivo. Theimportant differences noticed when transport ofcalcium was compared in similar segments ofintestine by perfusion in vivo and everted sac invitro were: (i) the greater efficiency of the gut fromvitamin D3-treated chicks in vivo, e.g. 38,ug. of45Ca transported in 30min. compared with 8,tg.in vitro (at lmM); (ii) after 30min. the amount ofcalcium found in the mucosa was always greaterin vitro than in vivo. This latter finding was due tothe large amount of 45Ca found in the muscle layerof the everted sacs. With everted sacs where theblood supply has been interrupted and the vesselscollapsed, it may be difficult for the 45Ca to pass tothe serosal compartment. The 45Ca found in themuscle layer may be calcium on its way from themucosal cells to the serosal surface. The accumula-tion of calcium in the muscle layer limits the useful-ness of the everted-sac technique for investigatingthe effect of various factors on calcium transport.An instance of the difficulty of interpreting resultscan be given when considering Table 5. The45Ca concentration in the muscle layer of sacsfrom rachitic chicks increased and that of thesacs from vitamin 13-treated chicks decreased ascompared with the non-inhibited controls. Theelevation of intracellular 45Ca concentration of

482 1965

Vol. 96 VITAMIN D3 AND CALCIUM ABSORPTION 483cells from rachitic chicks by iodoacetate shouldincrease the amount of 45Ca escaping via thereversible diffusion leak K2 (Scheme 1) to theserosal fluid. It was this 45Ca that probablycaused the rise in muscle 45Ca concentration andonly a small proportion found its way to the serosalfluid. Similarly, since less 45Ca was leaving theiodoacetate-inhibited cells from vitamin D3-treatedchicks than the corresponding control cells, therewas less 45Ca in the muscle layer of inhibited sacs.The location of 45Ca in mucosal cells loaded either

in vivo or in vitro was the same and it was mainlypresent in cell fragments that were centrifugeddown between 1000 and 3000g in 20% sucrose.What this fraction represents is notknown precisely,but it is most likely to be cell membranes and brushborders. This would be consistent with the idea ofpumps located at the cell surface, but the resultsdo not support the suggestion (De Luca, Engstrom& Rasmussen, 1962) that the mitochondria are themain site of vitamin Ds action.

Implicit in the model proposed is the idea thatvitamin D3 in some way inhibits the metabolicprocesses that drive pump P1. Evidence for aninhibition of metabolism by vitamin D3 has beenput forward in a series of papers (De Luca, Gran &Steenbock, 1957; De Luca, Gran, Steenbock &Reiser, 1957; Norman & De Luca, 1964). Both inkidney and in bone a block in metabolism betweencitrate and a-oxoglutarate was obtained withvitamin D3-treated rats. It is not known whethervitamin D3 inhibits citrate metabolism in theintestine of the chick, but it is difficult to believethis inhibition to be the only site of action ofvitamin D3, since the vitamin enhances calciumtransport in vitro under anaerobic conditionswhere presumably citrate metabolism is at astandstill.

REFERENCESAlbright, F. & Reifenstein, E. C. (1948). The Parathyroid

Glands and Metabolic Bone Disease, p. 125. Baltimore:Williams and Wilkins Co.

Bray, G. A. (1960). Analyt. Biochem. 1, 279.Coates, M. E. & Holdsworth, E. S. (1961). Brit. J. Nutr. 15,

131.De Luca, H. F., Engstrom, G. W. & Rasmussen, H. (1962).

Proc. nat. Acad. Sci., Wash., 48,1609.De Luca, H. F., Gran, F. C. & Steenbock, H. (1957). J. biol.

Chem. 224,201.De Luca, H. F., Gran, F. C., Steenbock, H. & Reiser, S.

(1957). J. biol. Chem. 228, 469.Greenberg, D. M. (1945). J. biol. Chem. 157, 99.Harris, L. J. & Innes, J. R. M. (1931). Biochem. J. 25, 367.Harrison, H. E. & Harrison, H. C. (1951). J. biol. Chem.

188, 83.Harrison, H. E. & Harrison, H. C. (1960). Amer. J. Physiol.

199,265.Krebs, H. A. & Henseleit,K. (1932). Hoppe-Seyl. Z. 210,33.Migicovsky, B. B. (1957). Canad. J. Biochem. Physiol. 85,

1267.Miller, D. & Crane, R. K. (1961). Analyt. Biochem. 2, 284.Nicolaysen, R. (1937). Biochem. J. 81, 122.Norman, A. W. & De Luca, H. F. (1964). Biochem. J. 91,

124.Parsons, D. S. (1963). Nature, Lond., 197, 1303.Rasmussen, H., Waldorf, A., Dziewiatkowski, D. D. &De Luca, H. F. (1963). Biochim. biophys. Ada, 75, 250.

Sallis, J. D. & Holdsworth, E. S. (1962). Amer. J. Physiol.203, 506.

Schachter, D. & Rosen, S. M. (1959). Amer. J. Physiol.196, 357.

Ussing, H. H. (1949). Ada physiol. 8cand. 19, 43.Waserman, R. H. (1964). Nature, Lond., 201, 997.Wasserman, R. H., Comar, C. L., Schooley, J. C. & Lenge-man, F. W. (1957). J. Nutr. 62, 367.

Wasserman, R. H. & Kallfelz, F. A. (1962). Amer. J.Physiol. 208, 221.

Webling, D. & Holdsworth, E. S. (1965). Biochem. J. (inthe Press).