Uptake and intracellular handling of iron from transferrin and iron chelates by mitogen stimulated mouse lymphocytes

tlhu-hh~.~k.,.~ ee IB.~hvsirc= ¢l~'t~. 1133 i It-~:'21 117 152

I ~7

,~i 19~)2 Elscvit:r St:icncc Publishcr,~ n . v All dl.,h=s re.,el-~cd fH~7.4HPq/'~2/$115.¢III

BBAMUR 13082

Uptake and intracellular handling of iron from transferrin and iron chelates by mitogen stimulated mouse lymphocytes A. D j c h a a n d J.H. Brock

IRe[rived I JIll~ 19ql)

Key wcwds: Transferrin: Ir*m-chelale; T cell pmlit~ratMn: Inlraccllul:tr irish di,,Irihulit~n: Ferrilin ~vnthe~.i~:( M,.lu.~' lyrrtphl~34¢)

The ability of lymphocytes to utilise iron from different sources has been investigated. Iron uptake from transferrin hy proliferating lymphocytes 8radually increased as saturation of the protein with iron was increased up to 100%, but rose sharply when addition of further iron resulted in the presence of nan-transferrin bound iron. Increasing the saturation of transrerrin with iron caused an increased rate of proliferation up to about 100% saturation hut when the level of iron present exceeded the binding capacity of the protein, prollferat[on decreased and at high levels of iron it was reduced below that seen in the absence of tran~.Terrin. Comparison of the degree of iron uptake from transferrtn and from iron chelators showed that the hydr~phili¢ chelator ferric nitrilotriaeetate (FeNTA) donated larger amounts nf iron to cells than did transferrin or the lipophilie chelator ferrie-pyridoxa! isonieotinoyl hydraznne (IFePIH), but did not promote proliferation, and when present in high amounts caused inidhition. In contrast, FePIH supported proliferation as efficiently as transferrln. In cells cultured with FeN'fA, iron was found predominantly in an Insoluble form while in the cells cultured with Fe-transferrin or FePIH the largest proportion of iron was found in the non-ferritin high molecular weight fraction, which probably represents iron in enzymes and other metabolically.important proteins, in no case did iron associated w~th ferritin exceed 15% of the total uptake, and the ceils showed no marked increase in synthesis of ferritin in response to ~=nyof the forms of iron. "l~ese results indicate that different forms of iron are handled in differtnt ways by l y m p h ~ t e s , and that iron delivered from hydrophilie chelates may be toxic and nat readily available for metabolic use. Lymphocytes appear to be poorly equipped to sequester excess iron in ferritin, and this may account for ahnnrmalities in the immune ~stem reported in patients with iron overload.

Introduction

The ability of lymphocytes to proliferate is closely linked to the supply of iron, it being needed in particular for the synthesis of iron-containing proteins and enzymes involved in DNA synthesis [1]. Both inadequate and excessive amounts of the metal can be deleterious [2,3]. Transferrin-bound iron appears to be the form in which the metal is normally supplied [4,1,5] and the ability of Tf to promote lymphocyte proliferation is closely related to its ability to donate iron to the cells

Abbreviations: NTA. r=itrilotriaeetale" h?drazune; Tf, tri:n.~tcrrin.

PIH. pyridaxal istmitxrtinnyl

Correspondence: A. Djeha. Deparlment of Immunology, Western Jr,fie'murk. Glasgow G l l 6NT, U~K.

[6]. However, there have been conflicting reports on whether iron sup~,li~d in other forms leads to inhibilion or enhancement of the lymphocyte response to mitogens, as some have reported that non-transferrinhound iron cannot support proliferation [7,4,5] while others have reported good proliferation [8]. The reasons for these discrepancies are not clear, but may he related to the type of iron carrier used, and the way in which different forms of iron are handled by the cells. There has been considerable investigation into the incorporation of iron into different intracellular iron compartmenls [9]. However, motet studies have been carried out using erythroid ceils, and as far as lymphot.~te~ are concerned thi,s a~ea of iron metabolism is almost unexplored, The few such studies of cells of the lymphomyeloid series [10,11] used Tf as the iron earlier and demonstrated that the uptake of iron is initially into an intraccllular non-[erritin compartment, followed by. a lime-dependent transfer oI" some iron to

148 fcrrltln. Howcver, in these studies the cells were cultured for only a very short time after iron administration, and there are no data concerning the steady-state intracellular distribution of iron in cultures incubated over a longer period, or of the fate of iron acquired from sources other than Tf. The work reported here has ~hcrefore examined the ability of iron in different forms and amounts to promote lymphocyte prolife;'ation in vitro, and has related this to the iron donating ability of the various compounds and the subsequent intracellular distribution of the metal. We have compared the effects of the low affinity hydrophilic chelator nitrilotriacetate (NTA) with that of pyridoxal isonieotinoyl hydrazone (P1H), a lipophilic chelator which can donate iron to erythroid precursors for use in haem synthesis [12] and of transfen-in-bound iron. Materials and Methods

Celt culture Lymph node cells from adult B A L B / c mice were cultured at 2- 10"/ml for 48 h in conical test tubes in serum-free RPMI-1640 culture medium supplemented with l m g / m l human serum albumin (Behrlngwerke, Hounslow, U.K.) and 1 ~ M 2-mercaptoethannl [41. Proliferation was induced by 2.5 # g / m l coneanavalin A (Sigma, Peele, U.K.). Human Tf (Behringwerke) at 0.625 # M (50 bcg/ml) was added when requiled and the appropriate saturation with iron achieved by prior addition of iron nitrilotriaeetate (FeNTA) to the stock solution [13]. Pyridoxal isonicotinoyl hydrazone (PIH) was kindly provided by Dr. P. Ponka (McGill University, Montreal). FePIH was prepared as described by Brock and Stevenson [14].

Measuremem of proliferation Proliferative responses were assayed after 48 h by placing 100 pl aliquots of the eel|s in a conical bottomed microtitre culture plate and pulsing with ['~H]thymidine (I ~Ci/wcli: specific activity 52 C i / m m o l ) for 4 h, The cells were then harvested and [3H]thymidine incorporation determined by seintillalion counting.

Iron uptake and intracellular distribution of iron Uptake and intraceIlular distribution of iron was investigated in proliferating lymphoeytes that had been cultured as described above in the presence of [-~'~Fe]transferrin, [~OFe]NTA or [S'~Fe]PIH, as follows: FeNTA was trace-labelled with [~qFe]eitrate (specific activity 1331 M B q / m g F e ) a n d was either used as such or added to Tf or PIH to give [SgFe]P1H or [S°Fe]Tf. The Fe/ehelator ratio was kept constant throughout all the experiments (1:4 for FeNTA and 1:2 for FeP1H) but the Tf concentration was maintained at

0.625/aM and tile saturation varied from 15% to 75%. After termination of the incubation period, the cells were washed three times with Hanks' solution and ~ F e incorporation determined in a gamma counter, For determination of intracellular iron distribution, the cells were lysed by adding 1 ml of I% Triton X-lO0 in PBS containing I mM desferrlo×amlne (DFO: Ciba Laboratories, Horsham, U.K.) to the pellet. The addition of D F O at this stage, which reaels with iron to form a highly stable complex, ensures that a~i chelatable iron is bound al the moment of cell lysis, hence reducing to a minimum potential movement of Iabile iron to other pools as a consequence of cell disruption, lntracellular distribution of iron was then determined by centrifugalion, immunoabsorption and ullrafillration of the lysate through a filter with a 10 000 Da cut-off, as described by Alvarez-Hern~ndcz et al. [LS] except that a multistage immunoabsorbent column in a 1 ml syringe barrel containing 50-75 tzl of each immunoabsorbent gel was used instead of individual columns. By this method eell-a~sociated iron was separated inte five fractions consisting of (i) transferrin-iron, (it) ferritin-iron, (iii) non-transferrim non-ferritin iron in a high molecular weight form, (iv) low-molecular weight iron chelated by desferrioxamine and (v) iron in an insoluble form. As mentioned earlier [15], these fractions do not necessarily conform rigorously to the forms of iron stated, but they provide a useful means of making comparative analyses of intracellular iron distribution,

Determination of ferritin Lymph node cells, cultured as described above, were lysed by freeze-thawing twice in distilled water to release intracellular ferritin. Estimation of ferritin in the lysate was carried out using a two-site immunoradiometric assay as described by Alvarez-Hernfindez et at.

[15].

Results

Effecl of iron saturation of transferrin on iron uptake and proliferation Cells were cuttured in the p r e s e n ~ of different iron saturatiors of [SgFe]transferrin and the amount of 59Fe associatea with file cells determined. Iron uptake increased with increasing Tf saturation up to 90% (Fig. 1) with about 10% of the available iron being acquired. Beyond this level, when the amount of iron present (added to the Tf as FeNTA) exceeded the binding capacity of the Tf present, the proportion of iron taken up increased, so that at the equivalent of 249% saturation cf Tf, the cells had acquired 32% of the available iron. Thus, non-transferrin-bound Fe (as FeNTA) seemed to be more readily taken up than Fe bound to Tf. Proliferation, assayed in parallel showed that at lower iron saturations (15-45%), the proliferative re-

149 Irc, o e o n e e n t r n t l o n (lt,%l) 0,36 ff+72 1.08 1.44 I,~ 2.16 2 :~2 2

4t) 353O •

i 25

rl.3'

i o.:

-e,,

20" IS"

¢,u

o o.+ N~

•

,

,

30

6~

90

,

.

,

.

IZO 150

,

+

,

,

lg(I 210 240

-'"

;-"

i

=7 --

Transferrin saturation [% t

sponse of the lymphocytes was lower than when Tf was fully saturated with iron. Thereafter, there was a sharp decline to below control level whetx the saturation of Tf exceeded 200%, suggesting that non~transferrin bound iron was inhibitory.

Proliferation of lymphocytes cldrured with FeNTA and FePIH Since FeNTA in excess of the binding capacity of Tf appeared to be inhibitory to lymphocyte proliferation, the effect of iron chelates was investigated in more detail. In the absence of Tf, a very low concentration (0.18 ~M) of iron as FeNTA had no significant effect on proliferation compared with control cultures with

.

"~ ~e~

3o+

i.~

20"

~

10"

0 -

5

0.'I lrnn concentration

(pM)

Fig. 3. [IH~hymidine incorporation by slimulaled mou~c I'+'mphoc3'tes cultured with Fc-pyridoxal imnicotino+~lhYdrazoneI.FePIH)at different clmceruration,, in the absence o f transferrin irne~m_+S.O., ,z = 4}, O l h e r details as Fig. IL.

no addition (Fig. 2). However, at higher iron concentrations a progressive inhibitory effect was seen, and at 1.6 ~M, all proliferation was abolished. Thus, FeNTA per ~ , rather than high levels of iron, appeared to be responsible for the inhibitory effect noted previously (Fig. 1). FeP|H at low concentration (0.27 tiM Fe) was also ineffective at enhancing proliferation (Fig. 3) but when the iron concentration was increased to 0.8-3,2 ~tM, proliferation increased, and was as good as that with 90% saturated FeTf (1.1 ~M iron), which was used as a positive control. Higher concentrations of FePIH were inhibitory.

Uptake of iron from transferrin and chelators by proliferating mo+tse lymphocyrcs The previous results suggested that there might bc differences in the degree of iron uptake from each source. The amount of iron taken up from identical concentrations of Fe added as either FeTE FeNTA or FePIH was therefore examined. Uptake of iron from both chelates was greater than from Tf (Fig, 4); at 0.18 p.M uptake from FeNTA was some 4.5-tlmes greater than from "If, while at 0.9 ,uM Fe, there was a 9-fold difference. This is in agreement with the increased uptake of Fe seen when the amount of Fe added as FeNTA exceeded the binding capacity of Tf (Fig. 1). Iron uptake from FePIH was lower than from FeNTA but stilI 1.5-fold higher than from T f a t both concentrations tested.

25" i

m

0 -

Fig. I+ Iron (e) and dwmid[ne { o ) uptak~ by stimulaled m,~,,,:e lyntphn~ytes c~llur~:d with traasterrin ;it d[t'ferent iron ~;tlurali~m.~ T h y m i d i n e uptake results are m~'an ¢ounls_+_ S+D, (a = 41 tn~m replicate wells: iron ~tptake results are sin$1v readings for parallel collures incubated with 5~Fe+ Results ate of ~ ~ep~esentati~rv ~xperim e a t v,'hich was r ep ea t e d three times.

_m

=

,~l. 0.2

0+¢

0.8

1 . 6

3.2

Iron concentration (pM) Fig. 2. ["H]'l'hymidille inc0rporutiQnby stimulated mou.~c lyrnphoceres e m t u r e d with Fe-nittiloltiacclal¢ { F e N T A ) at different coacen+ tralions in 1hi: absence o f transfcrrin ( m e a n :l: S,D.. ,l ==4). T h e solid b a r represents pr01ifer0fli0n in the absence of F e N T A . and the open b a r proliferation in the presence o[ 0.625 ~.M 9O%-saturaled Fetransferrin ( e v e control).

Imracellldar distribution of iron in proliferating mouse lymphocyle~ Since the carrier appeared to affect the amount of iron deIivered to the cells, it was of interest to determine the relative inlracellular distribution of iron taken up from these three sources. The major proportion of

150 .~

~'

0.4 ]

TABLE 1 httrrtc¢liuklr /~'rrilin [¢rrLv in C , n ,,1.~'ltmtdt/ted ~.J4z)rtw A'trpt:o~')tz'~ t'tdltrrcd fiJr 24 h i~t the f~r¢'~t'tt~'t~ o1" di(fc'rt'nt irt:n ~r.rrie'r~

0.3

Amount of iron added

Form ot iron

I'1

-

I 1.6-+- ~1 ~ "

FeTf

1 6 . 6 -- 3,fP

L I~

FeNTA

FePIH

I I FeNTA

FeTr

FePIH

C).9

[roR £al'l'ieF Fie-. 4. lm~ Uplal~c by 5timuhzted re,uSe lymphoqyte~ ¢ulturud with different into carriers at (L I~ a M Fe (stolid b;rr~)or f).L~~ M Fe (open bars), gcsults are of ~1 representative e×perintent, whicll wz.~ repealed ft~tnr limes. FcqL Fe IrunMurrim F c N T A , Fe-nilrih+tria,:¢tate; i:eP~H, Fe-pyridoxal isonicotJnoyl hydruzon¢.

iron taken up from FeNTA was found in the insoluble fraction which probably contains cellular debris and haemosiderin (Fig. 5). However, when Tf was present this fraction represented a much lower proportion and this was also the case for ceils cultured with FePIH. This suggests that much of the iron taken up from FeNTA is not used for cellular metabolism, The chelatable fraction comprised a fairly similar proportion of the Fe taken up from both FeTf and FeNTA and a slightly higher proportion of ICe derived from FePtH. although in absolute figures this fraction contained much more iron when FeNTA was the source. The proportion of non-tress(orris, non-ferritin high molecular weight-associated iron, which probably comprises mainly Fe bound to other proteins such as enzymes and thus represents iror, involved in cellular B0 ¢11

~- eo

~ 4o ~. 211

Ft

"If

Insol

HMW

Chel

Intracellular fraction

Fig= q Intraeellulardistribution of iron in ("on A :,timalatcd mouse nymphcJcyle~,eullured with 0.9 ,aM Fe as. Fe-transferrin(c~penburs), FeNTA (solid bars) or Pit] (hatched bac~). Results are from a [eprt:sentati~e experiment which was repeated four time~. FI, Ierritin; Tf. tran.sferrin; ln~L insolublema~efia[; t IMW. high mo[ecu]m wei/~ht huntress(orris, mm-[erritin fraction: Chel, chelae:able fraction.

FeTf FeNTA FcPIH

Ferric:in (ng/]l)" cells)

14,8 + 2.5

< 24.9+_4.8 lq.fl .L 5,It ] t 3+_2¢,

" mean + SD., n = 4.

metabolism, was much larger in the cells cultured with FeTf or FePIH than in cells cultured with FeNTA. This fraction was also greater, in absolute figures, in the cells cultured with FePIH or FeTf despite the overall higher uptake from FeNTA. The ferritin-iron and transferrin-iron fractions formed only a small proportion with all iron carriers. However, it should be noted that iron bound to Tf was found not only in cells incubated with FeTf but also in the cells incubated with ch~Iates in absence of any exogenous "If. There is some immunological cross-reaction between human Tf and mouse Tf (unpublished observation), and this might therefore represent iron bound t0 Tf synthesised by the proliferating mouse lymph node cells.

Intracelhdar ferritin let'els in proliferating mouse lymphocytes Since the proportion of iron incorporated into ferritin was found to be low, it seemed po.~sihle that lymphocytes contained relatively little ferritin. The effect of different amounts and forms of iron on intracellular levels of ferritin was therefore examined, The ferritin content of lymphocytes was generally low with all iron carriers, it being slightly higher in the presence of Tf, and lowest in the presence of FePIH (Table D. Furthermore, the cells showed only a very modest increase in intracellular ferritin levels when the concentration of Fe in the medium was increased from 0.18 to 0.9 .o.M indicating that stimulated mouse lymphocyte~ do not show any marked response to iron by increasing ferritin synthesis. It is of particular note that even the very high uptake from FeNTA (Fig. 4) failed t o increase ferritin levels, Discussion We have previously shown [4,14,16] that transferriniron or FePIH can enhance lymphocyte proliferation, but that iron delivered as FeNTA or iron citrate does

151 nat. The present work has extended these studies by examining the way in which iron from different sources is handled by lymphocytes, and provklcs some insight as to why some forms of iron stimuhde proliferatiun while others do not. Uptake of iron by lymphoeytes tYom Tf showed a near-linear relationship with iron saturation when this was < I(11)%, approx. IIl% of available iron being acquired (Fig. I~,. This is similar to the situation in other types of ceil such as cultured Chang cells [17]. liver slices, retieulocytcs and mueroph:~ges [18]. lIowever, in the present study it was found that when iron levels exceeded the binding capacity of Tf. the proportion of iron taken np rose sharply, suggesting that iron was now being acquired by a diffferent, and probably less controlled manner. This is supported by the fact that while the degree of proliferation incrcuscd with the amount of iron bound to Tf up to about ltli)f:~ saturation, further addition of iron resulted in a reduction in proliferation (Fig. 1). It seems likely that non-Tf-bound iron, in the form of FeNTA is cytotoxie, us relatively low levels of FcNTA were inhibitory to the baseline level of proliferation observed in the absence of Tf (Fig. 2). This indicates lhal there is no simple correlation between external iron supply and DNA synthesis by lymphocytes. When uptake of iron from FeNTA and FeTf were compared directly (Fig. 4~, it was found that the amount of iron taken up from FeNTA was much greater than from FeTf. These results are in accordance with our previous report that mouse lymphoeytes took up iron more rapidly from FeNTA and iron citrate e0molexes than from FeTf [16], and similar findings have been reported with Chang liver cells [19], which took up 30-fold more iron from FeNTA than from FeTr. In contrast, Taylor etal. [20] found that iron uptake from [~gFe]NTA added directly to cultures of mitogenstimulated human peripheral blood lymphocy~es was quantitatively very similar to that from [~'~Fe]Tf, However, their culture~ contained 5% serum, so it is likely that 'If in the serum mediated the uptake of [5'~Fe]. These results differ slightly from those of our earlier study, in which it was found that optimal proliferation occurred at 30-70% saturation of Tf, with a deerease at 70-100% [4]. The reasons for this discrepancy are not obvious but could be due to the fact that in this study preformed FeTf was added to the cultures~ while in the earlier work F'eNTA was added directly to Tf-containing cultures. This may have lead to the presence of free FeNTA in the cultures at higher iron levels, as binding is not instantaneous [21]. We have previously found that FePIH, unlike FeNTA supports good proliferation of Con A-stimulated mouse lymphocytes [141. Unlike FeNTA, FePIH is lipophilic and probably bypasses the Tf-TfR uptake system by delivering metabolically available iron directly through the

lipid bilai',:r [I21. Hoe, ever, levels of iron bound to PIH of more thzLn 2.q ~M inhibited cell proliferation, supgosling th:tt although iron bound to P1H may be able to bc utitised by the cell, its uptake is less ~cfi controlled lhan that of Tf-bound iron and ~my excess Qver that needed for biosynthetic processes becomes toxic. This is sug~orted by the fact that uptake of Fe from PIH. :dthough k~wer than t¥om FeNTA. was greater than from Tf. The intracctlulat distribution or iron acquired from the three carriers revealed differences that may rchtte to their differing degrees of uptake and ability ~o promote proliferation. In cells cu!tured with FeNTA iron was found predominantly in an insoluble fi~rm. while in cells cultured with FeTf and FePIH, this fraction represented only a small proportion of intracellular iron. This is in line with the report of White and Jacobs [lt)] who showed that iron delivered from FeNTA '~o Chang liver cells was found largely in the membrane pelfet and very little was in ~he form of [¢rri;in. The nature of the insoluble iron fraction is not certain. It may simply consist o[ FeNTA trapped in the membrane. However, iron in this form would ~robably be bound by DFO during the fractionation procedure. yet the proportion of cheIatable iron in cells incubated with FeNTA was not especially large. Alternatively. it could consist of haemosiderin, arising from saturation and then dcgrlid~allon of ferritin to which it was initially bound_ This also seems unlikely as only a small propor~:ion of Fc was found in ferritin. It seems more likely that iron in this fraction arises from the formation of polymeric complexes resulting from release of iron from a n d / o r catabolism of NTA within the cell. as reported by Nakamoto etal. [22]. In contrast the proportion of iron in the non-T/, non-ferritin high molecular weight fraction was highest. both in proportion and in absolute terms, in cells incubated with Tf and lowest in those incubated with FcNTA. This fraction probably contains mainly iron in enzymes and other molecules required for metabo|ic activity, and The results therefore ~,upport the suggestion that irm acquired from Tf is the form used most efficiently lor metabolic activity. The major prot~ortion of iron taken up from FePIH was also in ~his fraction. The fraction of intracetlular iron ehelatnb!e by DFO, which is thought to correspond to the intracellular Transit or labile iron pool [23], represented a relatively small proportion (10-30%~ of the iron in iymphoeytes_ Although the highest propOrtion was found in cclis cultured with FePIH, the amount of iron in this fraction was highest in absolute figures in cells cultured with FeNTA since these cells took up much more iron than the others. The low prot~ortion of iron found in ferritin, and the relatively small increase in cell ferritin levels in response to iron suggests that murin¢ lymphocytes have a limited ability to sequester unwanted iron

152 syntllesisin~ ferritin w h e n the m e t a l is p r e s c m e d in excessive a m o u n t s . T h i s a g r e e s with the findings o f P a t t a n a p a n y a s a t [24,25] w h o s h o w e d t h a t while P H A stimulation c a u s e d a significant i n c r e a s e in ferritin synthesis by h u m a n lymphocytcs, the p r e s e n c e of excess iron as F e N T A c a u s e d only a relatively m o d e s t ferritin increase, w h i c h in the case of the m o r e a b u n d a n t h e a r t - t y p e ferritin rapidly r e a c h e d a p l a t e a u as iron levels i n c r e a s e d . T h u s , a c o m b i n a t i o n o f excessive u p t a k e n-ore F e N T A a n d a failure to s e q u e s t e r this iron in fcrritin c o u l d explain the inhibitory cffcct o f FeNTA o n p r o l i f e r a t i o n , in c o n t r a s t , m u r i n e m a c r o p h a g e s showed n o evidence of toxicity w h e n exp o s e d to c o n c e n t r a t i o n s o f F e N T A up to 2ll{I-times g r e a t e r , a n d r e s p o n d e d with a v i g o r o u s increase in ceil ferritin levels [15]. Indeed, the possibility t h a t the m o d est r e s p o n s e to iron f o u n d in the p r e s e n t s t u d y was d u e to synthesis by c o n t a m i n a t i n g m a r i n e r n a c t o p h a g e s r a t h e r t h a n to l y m p h o c y t e s c a n n o t be ruled o u t . O n l y a small p r o p o r t i o n o f iron within l y m p h o c y t e s was b o u n d to Tf, a l t h o u g h interesting/y some w a s f o u n d even in cells that h a d b e e n c u l t u r e d in the p r e s e n c e o f c h e l a t o r s w i t h o u t a n y T f in the m e d i u m . This suggests t h a t m o u s e t y m p h o c y t e s m a y synthesise Tf, as h a s b e e n r e p o r t e d for h u m a n T cells [26]. W e are c u r r e n t l y investigating this possibility. In conclusion, this study s h o w s t h a t d i f f e r e n t i o n c a r r i e r s are h a n d l e d in different ways by ]ymphocytes, a n d t h a t the l y m p h o c y t e is a cell poorly e q u i p p e d to deal with excessive a m o u n t s of iron, p a r t i c u l a r l y if the m e t a l is p r e s e n t e d in a n o n - t r a n s f e r r i n - b o u n d f o r m w h o s e u p t a k e is poorly controlled. It has b e e n p r o p o s e d t h a t excessive u p t a k e o f n o n - t r a n s f e r r i n - b o u n d iron by liver cells m a y b e a m a j o r c a u s e o f iron l o a d i n g a n d s u b s e q u e n t liver d a m a g e in p a t i e n t s with iron o v e r l o a d [27], a n d the p r e s e n t s t u d y suggests t h a t a similar m e c h a n i s m m a y be b e h i n d s o m e o f the adverse affects o n tne i m m u n e system f o u n d in p a t i e n t s with iron o v e r l o a d [28]. Acknowledgement W e t h a n k Dr. P. P o n k a for s u p p l y i n g P I H .

References I Brock, J.ll. anti Mainou-Fowler. T. II983) lntrnanoL Today 4. 347- 35 I. 2 Dal/man. F,.R. (19871 Am, J. Clin. Nlitrilion 46, 329-334r 3 Good. M.F., Pnw¢ll, I,.W. and Ha[liOay, J.W. 09881 Blond gev. 2, 43-4q~ 4 Brock. J.H. (19g.II Immunology 43, 387-392. 5 Taylor. F,.G,. Soy~no, A . Rornatlo. E. and Lnyrisse, M. 0987) Tohokt~ J. Exp. Med. 153, 285-293. 6 Brock. J.H.. Mainou-FowMr. T. and Webster. L.M, (1'486) |m munology 57. 105-110. 7 Phillips. J.L and Azar[, P. ([q751 Cell. lmmunol. 15, 9a-99. a Tanno. Y. and T~kishima. T, (1982) TohoktJ J. Exp. Med. 136. 463-464. 9 Crlchmn, R.R, (1985) in F'roleins of Iron Storage and TranspOrt (Spik, G,, MQntrcuil. J., Crichlon, R.R, and Mui~uri=r, J,, cds.I, pp. q0-1 t0. Ehcviet, Amsterdam. Ill Born[oral. A, Young, S. and Williams. R. (lt~Sfi) Dr. J. Haematol. 62, 487-494. l[ Maltia, E., Josic, D., AshwelL G.. Klaesner, R. and Van Renswoudc, J. (1986} J. Biol. Chem. 261, 4587-4595. 12 Ftmka, P,, Sehulman, HM. and Wilezynska. A. (1982) Bioehim. P,iophys. Acta 718, 15]-13fi, 13 ~.ale.,,. G.W~ and Sehlabach. M.P-.. (1973) 2. Biol. Clem. 248, 3228-3232. 14 Brock, J'.H. and Stevenson, J. 09871 lmmureol. Leu. 15. 23-25. 15 AIvarez~Hem:indez, X.. Fe.lslein, M.V. and Brock, J.H. (19861 Biochim. Biophys. Aela 886. 214-277.2. 16 Brock+l.ll. and Rankia, M.C, {19gl) Immunology 43. 393-30& 17 Bailey-Wood, R.. While, G,P. and Jaeobs. A+ (197'~) Br, J. Exp~ Pathol. 58, 358-362. 18 Morgan. ErFL (1974) m Ion in Biocheraistry and M~d~eine (Jac~bg, A. ~nd Wonxood, M., cds.) pp. 3n-71, Academic F,rcss, New York. 19 White. Q,P. and lacobs, A. (1978j BioehJr~. Biophys. Acta 543. 217-225. 20 Taylor. P,G., Soyano, A.. Romann~ E. and Layrisse, M. (1988) Mictobiol. IHmunol. "~2.,949-955. 21 gar~]ski, E.t, and Pfine[o:lt,. J.V, (1977) in Proteins o[ Iron Metabolism :Bmwn, E.G., Alton, P.. Fielding, J. and Criehton. R,R., cds.L pp. 205-210. Prune and Strallon, New York. 22 Nakamolo, S.+ Yamanoi, S.. Kawataba, T., Sahira, Y,, Mort, M. and AwaL M. (1986) Bioehim. Biophys. Aeta 889, 1.5. 23 Jacobs, A, il9771 Blood 50, 433-439. 24 Pattanapanyasat, K., tlo~, T,G~ agd Jacobs, A. (1988) Br, J Haemalol, 69, 565~570. 25 F,attanapanyasat. K, (1989)Eur~ J. ltaematoL 43, 143-149. 20 Lure, J.B., Intante, A.J~, Makker, D.M., "Yang, F. and Bowman. B.H. 0986)J. Clin. Invest_ 77, 841-849. 27 Hershko, C. and P¢to, T.E.A (1987) Bt. J. Haematol 66, 149-I51. 2~ De $ousa, M. (19g,9) in Iron and Iglmunity~ Cancer and Infl~mmatlon (de Sousa, M. and Brock. t . H , eds.L !up. 247-258, Wiley. Chiehester