Variations of Human Blood Cell Zinc in Disease

Robert E. Fredricks, … , Kouichi R. Tanaka, William N. Valentine

J Clin Invest. 1964;43(2):304-315. https://doi.org/10.1172/JCI104915.

Research Article

Find the latest version: https://jci.me/104915/pdf Journal of Clinical Investigation Vol. 43, No. 2, 1964

Variations of Human Blood Cell Zinc in Disease *

ROBERT E. FREDRICKS, KouICHI R. TANAKA, AND WILLIAM N. VALENTINE (From the Department of Medicine, University of California Medical Center at Los Angeles, and the Wadsworth Hospital, Veterans Administration Center, Los Angeles, Calif.)

Variations in the amount of zinc in human Methods blood cells have been correlated with several dis- The methods of obtaining cells, preparing erythro- eases and morphological features. In 1949, Val- cyte- and leukocyte-rich suspensions, and measuring the lee and Gibson observed that in pernicious anemia acid-extracted zinc in a dithizone and carbon tetrachlo- erythrocyte zinc is elevated to a degree even ex- ride solution have been previously described (9, 12). ceeding what might be related to macrocytosis In most cases duplicate determinations were made on (1). A correlation of carbonic anhydrase activity each sample, and the duplicate values were averaged, so that a single result, corrected for erythrocyte or with erythrocyte zinc content was demonstrated leukocyte contamination, was recorded for each sample (2, 3). Subsequent reports concerning erythro- studied. cyte zinc have shown that it is decreased in the Leukocyte alkaline phosphatase was determined ac- fetus and newborn (3) and is increased in some cording to the method of Valentine and Beck (13). Standard methods were used to measure the hemato- patients with chronic leukemias and sickle-cell logical parameters, except that total blood cell counts anemia (1, 2, 4-8). The first evidence concern- were done with a Coulter counter and randomly ing leukocyte zinc abnormalities was reported in checked by visual counts in hemocytometers. 1949 by Vallee, Gibson, Fluharty, and Nelson The usual clinical and laboratory criteria were used (1, 4), who noted markedly decreased zinc in in establishing diagnoses, and each case was reviewed by leukocytes from patients with at least two of us, so that equivocal cases would be ex- chronic leukemias. cluded. In this report the term "acute leukemia" in- They also described a rise in leukocyte zinc to- cludes all those cases that might be classified as acute wards normal when subjects with chronic granu- or subacute leukemia. Included in the groups labeled locytic leukemia responded to therapy. We have as untreated are patients who have had no treatment for reported on the decreased amount of zinc in leu- their disease for a period of at least 6 months, as well as those who have never been treated. kocytes of patients with hepatic cirrhosis (9), The range and mean values for zinc, in each instance and high zinc levels in eosinophils have been de- recorded here, are the range and mean for all determina- scribed (10, 11). tions within the group. If serial values for each pa- This study was undertaken to evaluate the tient are averaged and the mean value then determined, levels of erythrocyte and leukocyte zinc in a the result does not vary significantly from that calcu- large group of patients with a variety of hemato- lated for all determinations. logical and nonhematological diseases and to ex- Results amine these cellular zinc levels in relationship to peripheral blood cell morphology, blood counts, Values in patients with leukernias. Leukocyte leukocyte alkaline phosphatase activity, disease and erythrocyte zinc determinations were per- state, and treatment status. Included in the formed on samples from 12 patients with chronic study are patients with acute and chronic leuke- lymphocytic leukemia, 22 with chronic granulo- mias, polycythemia rubra vera, myeloid meta- cytic leukemia, 3 with acute lymphocytic leukemia, plasia, megaloblastic and other anemias, eosino- 12 with acute monocytic leukemia, and 7 with philia owing to various causes, and pneumonia, and acute granulocytic leukemia. The results are tabu- patients in the postpartum state. lated in Table I. In all the groups of patients with leukemia the * Submitted for publication July 30, 1963; accepted range of values and mean values for leukocyte October 23, 1963. This investigation was supported by grants from the zinc are less than normal, with the lowest values U. S. Public Health Service, Parke, Davis & Co., and in chronic leukemia and acute lymphocytic leu- the Gladys F. Bowyer Fund. kemia and intermediate values in acute mono- 304 VARIATIONS OF HUMAN BLOOD CELL ZINC IN DISEASE 305

TABLE I Blood cell zinc values in leukemia

No. of deter- Leukocytes Erythrocytes mina- No. of Diagnosis tions patients Range Mean SD Range Mean SD

pg Zn/J10" pg Zn,/101" Normal 38 32 56.8-168 103.0 25.5 9.3-15.5 12.1 1.51 Chronic lymphocytic leukemia 15 12 0 - 78.7 52.2 20.0 7.7-15.9 13.0 2.10 Untreated 7 7 0 - 78.7 50.3 26.4 7.7-15.9 12.7 2.54 Treated 8 6 33.8- 71.3 53.8 11.8 10.3-15.5 13.3 1.97 Chronic granulocytic leukemia 36 22 17.8-130 58.7 22.2 7.0-18.5 13.3 2.65 Untreated 8 8 20.0- 64.3 48.4 15.7 10.4-17.3 14.0 2.13 Treated 28 19 17.8-130 62.1 22.8 7.0-18.5 13.1 2.77 Acute lymphocytic leukemia 4 3 34.3- 74.5 58.5 16.7 13.3-18.8 16.4 2.41 Acute monocytic leukemia 16 12 52.6-103 80.6 13.9 9.9-16.8 13.1 1.73 Acute granulocytic leukemia 7 7 43.9-122 94.4 26.6 7.4-13.2 11.8 1.84 cytic and acute granulocytic leukemia. If re- locytic leukemia r is minus 0.558, in chronic suits are evaluated within each group of patients, lymphocytic leukemia r is minus 0.254, in acute no correlation of leukocyte zinc with total or dif- granulocytic leukemia r is plus 0.522, in acute ferential leukocyte counts can be demonstrated monocytic leukemia r is plus 0.217, and in acute (Tables II and III). Correlation coefficients in- lynmphocytic leukemia r is plus 0.101. The inde- dicate the lack of correlation between total leuko- pendence of leukocyte zinc values and the dif- cyte counts and leukocyte zinc; in chronic granu- ferential leukocyte counts is further emphasized

TABLE II Blood cell zinc values in patients with chronic lymphocytic leukemia

Imma- Age Mature ture Race Leuko- Leuko- lympho- lympho- Granu- Erythro- Patient Sex Date cytes cytes* cytest cytesj locytesj LAP 1I cytes PCV¶ Treatment pg Zn/101 % % % pg Zn/101 A.B. 47, W, M 2/18 46.1 50,320 97.0 0.5 2.5 7.8 14.3 33 On prednisone T.B. 59, N. M 12/18 42.9 191,000 91.0 9.0 7.7 39 None J.D. 58, W, M 9/16 31.3 140,000 95.5 4.5 14.7 47 None R.D. 82, W, F 11/9 64.0 156,000 80.5 4.0 15.5 14.3 29 X ray to spleen 12/11 62.3 129,000 80.0 2.0 18.0 4.4 10.3 19 On chlorambucil H.H. 63, W, M 1/26 49.5 38,200 30.5 49.0 16.0 6.7 15.5 23 On 6-mercaptopurine H.K. 67, W. M 10/22 73.6 89,400 84.5 14.5 13.7 39 None J.L. 64, W. M 3/22 0 450,000 99.5 0.5 12.3 29 None F.P. 63, W. M 6/2 71.3 30,630 77.0 1.0 21.5 12.0 48 On prednisone J.R. 61, W, M 5/8 43.3 112,000 97.0 3.0 13.6 42 On triethylenemelamine 5/14 59.9 118,000 94.0 6.0 14.6 43 On triethylenemelamine V.S. 60, W, M 5/27 51.5 175,000 91.5 2.0 4.5 10.8 34 None 6/3 33.8 340,200 91.5 0.5 6.5 11.4 31 On prednisone and chlorambucil C.T. 62, N, M 5/21 74.1 31,000 67.0 29.0 13.8 42 None L.W. 65, W, M 2/18 78.7 35,900 89.5 10.0 15.6 15.9 29 Only transfusions for 10 months; previous X ray, triethylenemel- amine, chlorambucil

* Leukocytes per cubic millimeter in peripheral blood. t Percentage of mature lymphocytes in peripheral blood. $ Percentage of prolymphocytes and lymphoblasts in peripheral blood. § Percentage of all granulocytes in peripheral blood. Leukocyte alkaline phosphatase. ¶ Packed cell volume or hematocrit. 306 R. E. FREDRICKS, K. R. TANAKA, AND W. N. VALENTINE

0) "I .. 0 0 ol >

0. .0 U)0 .2 u . 4U) U1) 0.ri .- U c) r- 0 De 0% 0 0 ~' 00 C's. r.t 0 U)U)._O U M U CZ E %..0 0 U) C ^~ 0 _ U1) a) U) CZ z a Z ; 0.Q. (v.0 U O = 0 .0 0 00%0 0000 04 U O- 1.0~ bCd 4- 0 ; C- t-toO 0= 0 00::oC,4 0 4 CN Cs _,L C14 U) 4 U) 't t--: O6-me er° \6N oON

m m o 4 0~~~~~c .0 O O 8OZ o4 oo U) 0 0 O, %,o *~' 0. . cN ci c- co oo rz en - 000000 00 .4- N -4 _4V- :h> 0 O O O C) C) C) m tf cq 00 0% V o. _4 C- CO -4 cl _l\6 _; _; _ _li U)0 .)4.)0%ciuU ULn0 0 0 U) ) I00 U) o) U)U 0 0 e \0 '0 (Ii tt m a AtS

$ 06 t 0 U) 0v e to 111 4 c-.&14 t-: C.1i6 ., 4

b 0 ta4 Cd + 0 0 U) U) U) U) 0 U) U) U) 0 U) 0 0 0 ) 0) "m M ,I M C4 o C mo U -U)- 0O ¢ r. 00C U)

0 4-- C6 9 0 U) 0 U-) 0 U) 0 0D U-) 0n 00 0 O0 Cl) 100 e,00 t-. ~4 U)'6 0% ~4 ~, - U) 6 o6 Cd - - C-4 " - - C4- - N4 O 0) 00 - 0-o 0 t In 0 U) 0 ) 0 U) 0 00en -60 o o U)0 0 4 v0 U) -i 0-) U-6U) o0-400WC4 ein~U) 0 2 E e _~64

O O 8 0 0 0 0 0 0 - O0 0 0 0 0 0 0 0 0 00 0 0% 0 U) es 0 4- 4- C14 U) 6* 0 0 00 _~ 0 00 ci -_ '0 0'0 0 -N -4 - cd>U U)4 '0-N' -~ 0 00 o6 o~ o oo o . O a n 0 U U) oE so 9: ~m N _- I._ 0 00 0% 0 0 0 C- 0 _ 00 ) - - m U) 0- '0 o 4 oN t.. '0 C. N - 0. UX 00) U) b eio U) a.- -a.- a.- -. O-a Oa- s 4I- U) U) U) co c- - 0- 00'0 U) uC '4 U) 0 0 - - U) a -a- -a- - \0 - c,- - e- ) -4 U) U -0 U)

I- U)34 ) U) - -I -_ 3) 114 U) I-'10 It U U o4- U).0 C.)E

.) * 0 0.) -r -~ CZ VARIATIONS OF HUMAN BLOOD CELL ZINC IN DISEASE 307

Et CZ

-- U) U)~~~~~~4- - ~= U .C oC d C _ CZ

Coeqoo+Uo~~~~-U)i)o e U 0-U oCo0 _ 0 00 m ( m EC

I>U) u O U) ) E OU) U ) C

> i Q .dq rf %_ _ O tn- Ln00U)U) 0 U') U

U (N #)!o n (N ( Ne 5~) N :L

- c ) In r oo6

0 C')0 0 ) 0U) U)0 m U) O *a4-~~~~~~~~~~~~~~~~~~~~~~~~~1m O O OU: m A-% 0%t Co o ' N N '-o0% (NcN - 0

6 V 0 0 a Co0 U

.0 0a) U

0

CU %0 (N _ o o X m X N EDC \X g U) G ~o6UO ) yi,OciUOUoi%\CD O InOO ) ) ¢z ,) . I) W o W.f VU)) o o U) U) 0 00 -t00000 0C00 0 0

° °> O s~tl O O ° ° °°Co ° O* 0 °NO 0 6* 0 aU-tON' 0(')U) 0%--U--(Nt' ' O a V,= O C>_0 r CS 0 000 °n + h \ eq U ~~N N =~J 1O) o 4-0C)-

O0-X.*1.._.eoc e + - O O °

- 0 4) rN_oO00 t- U)uo \ N(N00oo C sK4 _ os w -o _;e (N - - \- ooU)(N0('0)''0U'oen- t _ t-\NU 0

U) _ C4 C3CU+0T0 308 R. E. FREDRICKS, K. R. TANAKA, AND W. N. VALENTINE by the similarity of zinc values for chronic leu- fan, but while he was receiving splenic irradiation kemia and acute lymphocytic leukemia and by the the zinc value was below the initial level. In fact that leukocyte zinc values for acute granulo- E.G. leukocyte zinc first declined with busulfan cytic leukemia are intermediate between normal therapy, but rose when good control was achieved values and those of chronic granulocytic leu- with this drug. L.H. demonstrated his highest kemia. leukocyte zinc value while in his best remission In the leukemic patients the erythrocyte zinc with X-ray therapy to the spleen. A.K.'s lowest values have a broader than normal range, and the leukocyte zinc value was noted when he was not mean values are higher than normal within every being treated and was "out of control," and the group except that of acute granulocytic leukemia. value rose when he responded to splenic X-ray The degree of anemia, total and differential leu- treatment. M.L. responded well to busulf an kocyte counts, and treatment status do not cor- therapy and had a concurrent elevation of his relate in any way with the erythrocyte zinc, and leukocyte zinc. W.P.'s highest leukocyte zinc no definite relationship to the mean corpuscular value occurred while he was having his best re- volume of the erythrocytes could be established. sponse to busulfan. The information for W.T. Values in patients with chronic lymphocytic is inconclusive. leukemia. There is no significant difference be- Although the generalization about the lack of tween the leukocyte zinc levels of the treated and correlation of leukocyte zinc with total leukocyte untreated groups with chronic lymphocytic leu- counts applies to the group with chronic granu- kemia. Although there is no definite correlation locytic leukemia, the highest leukocyte zinc val- of leukocyte zinc with the total leukocyte counts ues were found in the cases with the lowest counts, in this group, it may be significant that no zinc and the converse is also true. In the twelve de- was recovered from the leukocytes of the patient terminations done on samples of peripheral blood with the highest leukocyte count (450,000 per with total leukocyte counts of 20,000 per mm3 or mm3), whereas the greatest leukocyte zinc levels less, leukocyte zinc values ranging from 37.6 to were found in the cases that had the lowest total 130 ug per 1010 leukocytes were found, whereas leukocyte counts (Table II). the range was 20 to 66.6 Mug per 1010 leukocytes Values in patients with chronic granulocytic in the eleven instances when the total leukocyte leukemia. The data for the chronic granulocytic count was 100,000 per mm3 or greater (Table leukemia group are remarkable in that leukocyte III). This may not be an exception to our ear- zinc shows a rise with response to treatment, as lEer generalization, but rather another reflection

demonstrated by the difference in mean values of the effect of treatment on leukocyte zinc in for leukocyte zinc between the treated and un- chronic granulocytic leukemia. treated cases (Table I) and also by the individual Values in acute leukemia. A summary of find- responses of patients for whom serial determina- ings in each type of acute leukemia is given in tions were done (Table III). Patient P.A. had Table I. The acute lymphocytic and acute granu- the most remarkable rise of leukocyte zinc with locytic leukemia groups contain only a single value a good response to busulf an. R.A. had an in- for leukocyte zinc in an untreated patient; there- crease of leukocyte zinc with a response to busul- fore, no judgment of the effects of treatment on

TABLE IV Blood cell zinc values in polycythemia rubra vera and myeloid metaplasia

No. of No. of Leukocytes Erythrocytes determin- patients Diagnosis ations Range Mean SD Range Mean SD

j.g Zn/1010 jug Zn/10JO Normal 38 32 56.8-168 103.0 25.5 9.3-15.5 12.1 1.51 Polycythemia 17 13 59.0-232 108.1 44.4 5.5-14.6 11.3 2.59 Untreated 6 6 59.0-232 116.2 59.2 7.3-13.9 11.5 2.09 Treated 11 10 59.5-143 103.8 32.9 5.5-14.6 11.2 2.92 Myeloid metaplasia 8 6 42.3-108 68.8 22.1 11.9-19.8 14.9 2.79 VARIATIONS OF HUMAN BLOOD CELL ZINC IN DISEASE 309

* 9 0Do CZ O En 4uzL P0 CU 0.~~~~~~~~~~~~~~~~~C

) 0000 e Tu0+0> d v U)U0 CU ~~o~~~~~ ~ ~ SEE~~~~ o C0 0 H to+\0m 0E O~css 0. 0eoiboo. oU)c oOo oNoOo f

0%0%t-.U--0%~~~~~- 00 00 0 0-0% 0% ooC En\0

u 0o~uooiq \0000omoon'o

0 0 0 001)- o0~ 0 00~00 U1 cd .0 V 0_eru%0: U) 0000 U.) -&)00'-jUj ") 0%") 00 \0 e u 0~~~~~0 oo14: oo.. o I',. o o1 o2 o o o o 00 U 040 0U) .0v11).0 a)") 11)011)0 0 -~~~~--o W) 0CN~~~~~~~1 rq 0 t cq 6 ot-~~~~~ 0m1' a, CZ~M

E i --t 0C 014.C:00. 00.-.IO. t00.Wrq. . . . .tU") .a, 00. t0.ooC40 e a190.)99 0 1 0 0 E- 00cs~m-Ln m __:Ne_-W) u0

0gt- r- 00 0,, 0 0,0r-C6r ON O e N 0 0_ O. cd ~e~U~e1 (1) 00~ ~ ~It ~c)uen ~ u N~~"-- -1)-CN 021 g~~~~~~~~~Q0ai 00LP)0 in0dWt0

C ;

8

d4- m) 01s) in1c1 0 e) 0000 W) 1.0 U% 0%i2) 0 U0 xs0 - 1-4 -0 X0:CX Cm - 4-b -6 0 0 To 000 0 000 n 0 o O ooo)0000 00u) 0000 00 I- I too0 - CNo4 r1I I- Uot-t-00\0 00o0 t-I0t\ \0 It (1) -6 a C)ooO O o Q o otnoocoo OoQ O o O 0 0 0* Uo0 a) CU V %011)4~och'0% '-' 0 11)J'N (1) t%0C4m%0 tCN 1)t U-to 00N O0 00l)In 0.. 04N a 0%%0000 0+ 0%0 0% 00010000 11)0% C000 O 0 1.4) 00 o 000o 0 eo U-n U- (I 000\0It C>000 :a tol

C14 ) e2 11) 0C-) N _-C- o%-% _-- * 0 U'20 C'-o C14~-0o~-e0 ~N0 ~ ~c-bo ~ C oooe 0%-+

Ce 'e DCU -- U)X ------U) oCu x I- C1 \ (: -

-oU.) 333~~~~~~C3CZ 333 B 3 P0 -5S oon,Ac 4:>oo csoC5~~u) N N ~~~~~~~~~~-oY _ of 310 R. E. FREDRICKS, K. R. TANAKA, AND W. N. VALENTINE leukocyte zinc can be made in these cases. Studies MCV of 124.4 ju8 is associated with the highest on cases of acute monocytic leukemia included de- erythrocyte zinc value of 19.8 jug per 1010 eryth- terminations of blood cell zinc of four patients rocytes as seen in Table V. who had never received treatment for their dis- Zinc values related to leukocyte alkaline phos- ease; however, the data do not show any re- phatase. In many of the cases of leukemia, poly- lationship of blood cell zinc to treatment. cythemia, and myeloid metaplasia, leukocyte al- Values in polycythemia rubra vera and mye- kaline phosphatase activity was measured on loid metaplasia. Thirteen patients with polycy- portions of the blood samples obtained for zinc themia rubra vera and six who had myeloid meta- determinations (Tables II, III, and V). The plasia are included, and the results are shown in results clearly show that levels of leukocyte alka- Table IV. In all but one of the seventeen de- line phosphatase and values for zinc in leukocytes terminations done on samples from polycythemic are completely independent of each other. patients, the leukocyte zinc values are normal. Values in anemia, eosinophilia, pneumonia, and The remarkably elevated value of 232 u.g of zinc postpartum cases. Three patients with megalo- per 1010 leukocytes was found in a blood sample blastic anemia (two due to folic acid deficiency from an untreated patient when he had a packed and one due to pernicious anemia) were studied. cell volume of 82%o (Table V). Four days later, Pretreatment blood samples were analyzed in each after many phlebotomies had reduced this to 76%o, case, and in each there was a markedly elevated the leukocyte zinc had fallen to 167 ,ug per 1010 erythrocyte zinc value. In the two cases of folic leukocytes. All the other values for the poly- acid-deficiency anemia in which posttreatment cythemia group fall into the normal range, have blood samples could be obtained, the erythrocyte a normal mean value, and show no relationship to zinc decreased as MCV declined and the packed total (r, - 0.297) or differential leukocyte cell volume and hemoglobin increased (Table VI). counts, packed cell volume, or treatment status. Sufficient leukocytes for zinc analysis could be In the cases of myeloid metaplasia leukocyte obtained only from patient H.W. after his third zinc is significantly decreased, and, again, with- week of folic acid therapy. The first leukocyte out relationship to total leukocyte counts (r, zinc value was significantly low, but the leukocyte 0.333), differential leukocyte counts, or hemo- zinc rose into the normal range after an additional globin levels. The available data do not permit week of treatment, when the erythrocyte zinc evaluation of the effects of treatment on zinc value had fallen to normal. levels. Patients with paroxysmal nocturnal hemo- The erythrocyte zinc levels in both polycythemia globinuria had normal erythrocyte zinc in the rubra vera and myeloid metaplasia seem to show four cases studied. Sufficient leukocytes for a direct relationship to mean corpuscular vol- study could not be obtained from these patients umes (MCV) of the erythrocytes. The highest because of the characteristic low leukocyte counts.

TABLE VI Blood cell zinc values in megaloblastic anemias

Age Race Leuko- Erythro- Leuko- Diagnosis Patient Sex Date Treatment cytes cytes PCV* MCVt cytes$ pig Zn/ pg Zn/ 1010 1010 Folic acid deficiency M.O. 38, W, F 4/22 None 19.7 27 120 6,200 Folic acid deficiency M.O. 5/9 On folic acid 15.3 33 124 5,300 Pernicious anemia F.R. 69, W. F 7/10 None 27.8 22 7,460 Folic acid deficiency H.W. 41, W, M 7/10 None 26.8 14 175 9,310 Folic acid deficiency H.W. 7/20 On folic acid 21.0 25 135 6,190 Folic acid deficiency H.W. 8/3 Oni folic acid 35.4 16.7 33 111 5,600 Folic acid deficiency ll.W. 8/10 On folic acid 75.1 12.7 34 110 6,350

* Packed cell volume or hematocrit. t Mean corpuscular volume of erythrocytes. I Leukocytes per cubic millimeter of peripheral blood. VARIATIONS OF HUMAN BLOOD CELL ZINC IN DISEASE 311 Eleven patients with other types of anemia Attempts have been made to correlate variations (two t halassemia muinor, three sickle-cell anemnia, inl letnkocyte zinc levels with the differential leuko- ()nC SC disease, one fava belan sensitivity anemlia cyte connts, the degree of inmmaturity of the leti- in renission, two idiopathic acquired hemolytic kocytes, the total leukocyte counts, and various anemias with remote splenectomies, one congeni- modes of therapy used in the diseases studied. tal hemolytic anemia with remote splenectomy, NVolff (11) has reported that eosinophils con- and one due to hereditary spherocytosis) had tain large amounts of zinc, and the histochemical evaluations of leukocyte and erythrocyte zinc, but studies of Mager, McNary, and Lionetti (10) no definite abnormality was suggested by the confirm this characteristic; however, the values results. found in this study do not show any correlation Seven patients with eosinophilia (ranging from of zinc content with the eosinophil counts in a 12.5 to 73.5%, and associated with parasitic in- group of seven patients with eosinophilia ranging festations, collagen diseases, drug reactions, and from 12.5 to 73.5%o. Extensive tables in this Loeffler's syndrome) were studied. Only one paper demonstrate further the great variations in zinc value was outside the normal range, a low differential cell counts present in cases with simi- value of 47.1 ug of zinc per 1010 leukocytes in lar leukocyte zinc values. For example, there is the patient with 73.5% eosinophils. The leuko- a striking similarity between the zinc levels in cyte zinc did not correlate with the differential chronic lymphocytic leukemia, where leukocyte leukocyte count or other measured values. A samples contain almost pure lymphocytes, and in patient with 58.5% eosinophils had 101 Ftg of chronic granulocytic leukemia, where leukocyte zinc per 1010 leukocytes, and another with 12.5% suspensions contain variable percentages of neu- eosinophils had 59.9 pg of zinc per 1010 leuko- trophilic forms, eosinophils, and basophils (most cytes. strikingly, 76%o basophils as shown for D.N. in Six patients with leukocytosis owing to pneu- Table III). monia had normal blood cell zinc values, except Wolff (11), Candura and Candura (14), and for the one patient who also had cirrhosis. In de Nicola and Candura (15) have offered evidence the latter, leukocyte zinc was distinctly low, 29.2 that the leukocyte zinc level varies with cellular pg of zinc per 1010 leukocytes. maturity, being lowest in the most immature cells. Leukocyte and erythrocyte zinc values on sam- Wolff based his conclusion on the evaluation of ples from one woman during labor and from cases of chronic leukemia, and the Italian work- six during the immediate postpartum period ers included just one example of acute leukemia were distributed in the normal range. among the seven cases of leukemia that they used in graphing this relationship. A relationship be- Discussion tween leukocyte zinc and cellular maturity is not Our studies (6) and those of Wolff (11) and apparent from the data detailed in Tables II, III, of Dennes, Tupper, and Wormall (7, 8) have and V, and the most obvious argument opposing shown again that there is a decrease in the zinc this concept is that the leukocyte zinc levels in content of leukocytes from persons with chronic acute granulocytic and acute monocytic leukemias lymphocytic and chronic granulocytic leukemias, exceed those found in cases with chronic leukemia, as originally observed by Gibson and associates as shown in Table I. (4). In addition, this report points up that a) An inverse relationship between the total periph- leukocytes of patients with acute lymphocytic eral leukocyte count and leukocyte zinc content in leukemia and myeloid metaplasia also have a low chronic lymphocytic leukemia has been reported level of zinc, whereas the leukocytes in acute by Dennes and associates (8). The data of this monocytic and acute granulocytic leukemia con- study do not reveal a similar clear-cut association, tain about 80 to 95% of the normal amount of although leukocyte zinc values in chronic lympho- zinc, b) the leukocytes in polycythemia vera and cytic leukemia and also in chronic granulocytic leukocytosis contain zinc in normal amounts, and leukemia are lowest in the cases that had the high- c) the leukocytes in megaloblastic anemia may be est leukocyte counts, and greatest in the cases zinc deficient. with the lowest total leukocyte counts. There is 312 R. E. FREDRICKS, K. R. TANAKA, AND W. N. VALENTINE no correlative association between total leukocyte ings and also those of Dennes and associates (8), counts of peripheral blood and leukocyte zinc who observed that the range of erythrocyte zinc values in the various other groups presented here, values is much greater in chronic leukemic sub- or in the previously reported study of patients jects than in normal subjects. In addition, we with cirrhosis (9) ; the maximal correlation co- have found elevated erythrocyte zinc levels in efficients are plus 0.522 and minus 0.558. patients with acute lymphocytic leukemia, acute As seen in Tables I and III, this study confirms monocytic leukemia, and myeloid metaplasia. the reports of Gibson and associates (4) and We have previously commented on the broad Wolff (11), who noted that leukocyte zinc con- range of erythrocyte zinc levels in patients with centrations rise in cases of chronic granulocytic cirrhosis (9) and on the possible correlation of leukemia which respond to therapy. A similar these values with the MCV of the erythrocytes; effect of therapy on leukocyte zinc could not be a similar correlation is suggested by the data for demonstrated for the patients with chronic lym- polycythemia vera and myeloid metaplasia tabu- phocytic leukemia, and insufficient information lated in Table V. No other relationship of eryth- is available for such an evaluation of the effects rocyte zinc to hematological determinations, dis- of therapy in the other patients with leukemia ease state, or treatment status has been detected. and myeloid metaplasia. Although only one case The direct relationship of erythrocyte zinc has been studied, it seems clear that there was a levels to carbonic anhydrase activity of the eryth- rise of leukocyte zinc f rom a low to a normal rocytes was first described by Vallee, Lewis, level, as the patient with megaloblastic anemia Altschule, and Gibson in 1949 (2). This close owing to folic acid deficiency responded to treat- association has never been discounted; however, ment. the study of Dennes, Tupper, and Wormall (19), In 1952, Hoch and Vallee (16) described the in which they measured radioactive zinc65 up- extraction of a zinc-containing protein from leu- take by blood cells, has demonstrated that there kocytes, which they estimated accounts for 80%o is a portion of erythrocyte zinc which exchanges of leukocyte zinc. The biological role and bio- freely with the body pool of zinc and is apparently chemical significance of this protein, and, indeed, not complexed in any intimate chemical union of leukocyte zinc in general, have not been de- with proteins of the cells, whereas the zinc of termined. It is assumed that this zinc metallo- carbonic anhydrase is a closely held constituent protein of leukocytes is an enzyme or group of of that enzyme and cannot be dialyzed out of the enzymes; however, Vallee (17) states that there cell even with potent zinc chelating agents (8). is no correlation of zinc levels and alcohol de- It is not likely that easily exchanged or "la- hydrogenase, carboxypeptidase, lactic dehydrogen- bile" zinc could contribute to the increased con- ase, or rhodanese activity in human leukocytes. centration of erythrocyte zinc reported in certain Leukocyte alkaline phosphatase, which is, most diseases. In the methods used in determining our likely, another zinc metalloenzyme (18), varies data and that of other published reports concern- in activity without relationship to leukocyte zinc ing increased erythrocyte zinc, the "labile" por- levels (9), as can be seen from the data pre- tion of zinc in erythrocytes is probably removed sented in this report. Other leukocyte enzymes during the cell washings (8) that precede the that have been studied in this laboratory, includ- analysis of the zinc content. The measured ing glucose-6-phosphate dehydrogenase, fumarase, erythrocyte zinc probably represents the "stable" aconitase, arginase, and lactic dehydrogenase, have zinc, the protein-bound zinc, which is incorporated shown patterns of activity that would not corre- in the carbonic anhydrase molecule. Previously late with the known variations of leukocyte zinc. published reports have shown that serum zinc Previously published reports have described levels are reduced in the clinical situations in abnormalities of erythrocyte zinc levels, specifically which we find increased erythrocyte zinc values elevations in pernicious anemia with return to (20). This fact would lead one to postulate normal as remissions are induced and high values that the "labile" zinc in these erythrocytes would in patients with chronic leukemias (1, 2, 4-8). decrease, because of the tendency for the "la- The results reported here substantiate these find- bile" zinc of the cells to be equilibrated with the VARIATIONS OF HUMAN BLOOD CELL ZINC IN DISEASE 313 Diagnosis WBC Zn* RBC Zn Serum Zn

Folic acid deficiency anemia 30%o 4, 4% Chronic lymphocytic leukemia 50% 4% Chronic granulocytic leukemia 60% 4% Acute lymphocytic leukemia 60% Myeloid metaplasia 70%6 4% Cirrhosis 70% N Acute monocytic leukemia 80% 4 Acute granulocytic leukemia 95% N Polycythemia vera 0O0% to + N to v N to 4% Anemias (non-megaloblastic) '100% N Eosinophilias 100%0 N Leukocytoses OOo N 4, Women at parturition 100% N

*WBC Zinc expressed as approximate percentage of normal mean value N = normal value = value greater than normal mean value

4, = value less than normal mean value

FIG. 1. SUMMARY OF THE RELATIONSHIPS OF LEUKOCYTE, ERYTHROCYTE, AND SERUM ZINC LEVELS IN VARIOUS DISEASES. body zinc pool, as most immediately reflected in long-standing disturbances in serum zinc. In the erythrocyte's plasma environment. cirrhosis serum zinc is depleted apparently be- Many of the findings discussed in this paper are cause of the associated zincuria (21). Increased graphically represented in Figure 1, which il- binding of zinc by erythrocytes may be responsi- lustrates that there is a reciprocal relationship ble for reducing serum zinc in cases of megalo- between leukocyte and erythrocyte zinc levels. blastic anemia, chronic leukemia, acute lympho- Erythrocyte zinc tends to be increased in those cytic leukemia, and myeloid metaplasia. Decreased clinical states in which leukocyte zinc is de- serum zinc might be expected to be reflected first creased, with the notable exception of the cases in the leukocytes, because they have a rapid turn- of cirrhosis in which the average erythrocyte over and also because they ordinarily take up zinc value is normal even though leukocyte zinc and bind 30 to 75 times more zinc than erythro- is decreased (9). Greatly elevated leukocyte cytes. Cases of polycythemia rubra vera may zinc was found in only one instance, in a patient provide us with a model for defining these inter- with polycythemia vera (S.B. in Table V), and relationships, since an occasional florid case has his erythrocyte zinc was reciprocally decreased. a reduction of erythrocyte zinc and an increase It has also been reported that serum zinc may be of serum zinc that may be responsible for the increased in some cases of polycythemia vera elevation of leukocyte zinc. (20). This single case then gives complementary strength to the many examples of a reciprocal Summary relationship of leukocyte and erythrocyte zinc noted among the groups with decreased leukocyte 1) Leukocyte zinc values for 46 patients with zinc. various acute and chronic leukemias were found The abnormal leukocyte zinc levels may reflect to be less than normal, and the lowest values 314 R. E. FREDRICKS, K. R. TANAKA, AND W. N. VALENTINE were noted in cases of chronic leukemia and acute References lymphocytic leukemia. 1. Vallee, B. L., and J. G. Gibson. The zinc content 2) The rise of leukocyte zinc content with re- of whole blood, plasma, leukocytes and erythro- sponse to therapy in chronic granulocytic leu- cytes in the anemias. Blood 1949, 4, 455. kemia is confirmed. 2. Vallee, B. L., H. D. Lewis, M. D. Altschule, and J. G. Gibson II. The relationship between car- 3) Leukocyte zinc was normal in 16 and ele- bonic anhydrase activity and zinc content of vated in one determination on samples from pa- erythrocytes in normal, in anemic and other patho- tients with polycythemia rubra vera, but sig- logic conditions. Blood 1949, 4, 467. nificantly decreased in six patients with myeloid 3. Berfenstam, R. Studies on blood zinc: A clinical metaplasia. and experimental investigation into the zinc con- tent of plasma and blood corpuscles with special 4) In one patient with megaloblastic anemia, reference to infancy. Acta paediat. (Uppsala) leukocyte zinc rose from a low level into the nor- 1952, 41, suppl. 87. mal range with response to therapy and subsi- 4. Gibson, J. G. II, B. L. Vallee, R. G. Fluharty, and dence of the anemia. J. E. Nelson. Studies of the zinc content of the leucocytes in myelogenous leukemia. Union intern. 5) Leukocyte zinc was normal in six patients clin. contra cancrum. acta. 1950, 6, 1102. with leukocytosis due to pneumonia. 5. Talbot, T. R., Jr., and J. F. Ross. The zinc content 6) Leukocyte and erythrocyte zinc levels were of plasma and erythrocytes of patients with per- normal in seven women during labor or the im- nicious anemia, sickle cell anemia, polycythemia mediate postpartum period. vera, leukemia, and neoplastic disease. Lab. In- vest. 1960, 9, 174. 7) Leukocyte zinc levels could not be correlated 6. Fredricks, R. E., K. R. Tanaka, and W. N. Valentine. with eosinophilia, basophilia, or other variations Leukocyte and erythrocyte zinc in blood dyscrasias in differential leukocyte counts or with total leu- (abstract). Clin. Res. 1960, 8, 209. kocyte counts or with the degree of leukocyte 7. Dennes, E., R. Tupper, and A. Wormall. Zinc con- immaturity. tent of erythrocytes and leucocytes of blood of normal and leukaemic subjects. Nature (Lond.) 8) The level of leukocyte alkaline phosphatase 1960, 187, 302. activity shows no relationship to leukocyte zinc 8. Dennes, E., R. Tupper, and A. Wormall. The zinc levels. content of erythrocytes and leucocytes of blood 9) Erythrocyte zinc content is often increased from normal and leukaemic subjects. Biochem. in patients with chronic leukemias, acute lympho- J. 1961, 78, 578. cytic leukemia, acute monocytic leukemia, and 9. Fredricks, R. E., K. R. Tanaka, and W. N. Valentine. Zinc in human blood cells: normal values and ab- myeloid metaplasia. normalities associated with liver disease. J. clin. 10) As reported by others, we found that eryth- Invest. 1960, 11, 1651. rocyte zinc is increased in megaloblastic anemia 10. Mager, M., F. F. McNary, Jr., and F. Lionetti. The and declines with response to therapy. histochemical detection of zinc. J. Histochem. 11) In some cases the erythrocyte zinc con- Cytochem. 1953, 1, 493. 11. Wolff, J. P. Untersuchungen zur Pathophysiologie tent seems to correlate with the mean corpuscular des Zinkstoffwechsels. Klin. Wschr. 1956, 34, 409. volume of the erythrocytes, but in most cases 12. Fredricks, R. E., K. R. Tanaka, and W. N. Valen- shows no correlation with the degree of anemia tine. A method for measuring zinc in leucocytes or other factors. and erythrocytes. Analyt. Biochem. 1961, 2, 169. 12) There does seem to be a reciprocal rela- 13. Valentine, W. N., and W. S. Beck. Biochemical studies on leukocytes: (1) Phosphatase activity tionship between leukocyte and erythrocyte zinc in health, leukocytosis, and myelocytic leukemia. levels in various disease states, and the possible J. Lab. clin. Med. 1951, 38, 39. implications of this observation are discussed. 14. Candura, F., and M. D. Candura. Ricerche sue con- tenuto in zinco del siero di sangue umano in con- dizioni normali e pathologiche. Prog. Med. 1957, Acknowledgments 13, 801. The authors are especially grateful for the skillful 15. De Nicola, P., and F. Candura. Rapporti tra im- technical assistance of Miss Marylu Mattson, Mrs. Julie maturita cellulare e zinco leucocitario nelle leucosi. Wittenberg, and Mrs. Eva Szentvari. Boll. Soc. ital. Biol. sper. 1958, 34, 713. VARIATIONS OF HUMAN BLOOD CELL ZINC IN DISEASE 315

16. Hoch, F. L., and B. L. Vallee. Extraction of zinc- zinc in blood; transport of zinc and incorporation containing protein from human leukocytes. J. of zinc in leukocytes. Biochem. J. 1962, 82, 466. biol. Chem. 1952, 195, 531. 20. Vikbladh, I. Studies on zinc in blood. Scand. J. 17.-Vallee, B. L. Biochemistry, physiology and pathol- clin. Lab. Invest. 1951, suppl. 2. ogy of zinc. Physiol. Rev. 1959, 39, 443. 21. Vallee, B. L., W. E. C. Wacker, A. F. Bartholomay, 18. Trubowitz, S. Isolation, purification, properties of and F. L. Hoch. Zinc metabolism in hepatic dys- alkaline phosphatase from human leukocytes (ab- function. II. Correlation of metabolic patterns stract). Blood 1960, 15, 419. with biochemical findings. New Engl. J. Med. 19. Dennes, E., R. Tupper, and A. Wormall. Studies on 1957, 257, 1055.