- Email: [email protected]
Serum-levels of lactoferrin, lysozyme and myeloperoxidase in normal, infection-prone and leukemic children
121
Clinica Chimica Acta, 136 (1984) 121-130 Elsevier
CCA 02734
Serum-levels of lactoferrin, lysazyme and myeloperoxidase in normal, infection-prone and leukemic children Per Venge a, Tony Foucard Departments
b, J&n Henriksen Anders Kreuger b
‘, Lena
H&ansson
a and
of” Clinical Chemistry and b Pediatrics, University Hospital, Uppsala (Sweden) and ’ Department o/Pediatrics, University Hospital, A’rhus (Denmark) (Received April 15th; revision August 18th, 1983) Key words: Blood ceil; Turnover; Granule prctein; Radioimmunoassay
Serum levels of lactoferrin, lysozyme and myeloperoxidase have been established in 31 healthy children. On average, serum lactoferrin was 330 pg/l, serum lysozyme 1638 pg/l and serum myeloperoxidase 174 pg/l. Serum myeloperoxidase was, on average, significantly higher in children than in adults (p = O.Ol), whereas serum lactoferrin and serum lysozyme were equal to those of adults. In a group of infection-prone children (n = 31), both serum lactoferrin and serum myeloperoxidase, but not the serum lysozyme levels, were significantly lower (p < 0.001 and p = 0.002, respectively) than those of the reference children in spite of normal intracellular contents and even somewhat higher peripheral blood polymorphonuclear counts. Based on the assumption that serum lactoferrin and serum myeloperoxidase reflect turnover and activity of neutrophil granulocytes, the findings could suggest reduction in these respects and could be one contributing factor to the high infection propensity of these children. Serum levels of .the three proteins have also been measured in 10 children with suspected or various forms of manifest leukemia. It is suggested that the levels reflect turnover and stage of maturation of the myeloid and monocytic cells and could, therefore, aid in the understanding and diagnosis of these diseases.
Correspondence and reprint requests to: Per Venge, MD, Department of Clinical Chemistry, University Hospital, S-750 14 Uppsala, Sweden. 0009-8981/84/$03.00
0 1984 Elsevier Science Publishers B.V.
122
Introduction Turnover and activity of leucocytes may be estimated by serum or plasma determination of leucocyte-derived proteins [l-3]. Lysozyme has been most extensively utilized for this purpose and in serum is probably derived from both neutrophils and monocytes/macrophages [1,4]. The relative contribution from these two sources is, however, uncertain. In the healthy subjects, a correlation with the blood neutrophil count is obtained [4], but in chronic inflammatory diseases such as sarcoidosis or Crohn’s disease [5,6], the high serum lysozyme levels most likely reflect monocyte/macrophage activity. In monocytic leukemia, serum lysozyme determinations are used to establish the diagnosis [7] and to estimate the leukemic cell mass [l]. Lysozyme is eliminated from the circulation through glomerular filtration (GFR) [8] and serum lysozyme levels are consequently raised when there is a reduction in GFR [8,9]. Lactoferrin is an iron-binding protein contained within the secondary (specific) granules of neutrophils [lo] and serum/plasma levels of this protein are probably specific markers of neutrophil turnover and activity [2,3,11]. Serum/plasma levels of lactoferrin in the healthy correlate with the number of blood neutrophils [2]. One report [12] has shown a 50% reduction in the plasma levels in women under the age of 50 years, probably due to the reduction of lactoferrin within the circulating neutrophil [13]. Monocytes/macrophages seem to have specific receptors for lactoferrin [14]. The catabolism may, therefore, be dependent on the availability of these structures. An uptake of lactoferrin in the liver also points to a dependence on liver function for the elimination of lactoferrin [3]. Myeloperoxidase is contained within the primary (azurophil) granules of neutrophils but also to some extent within the monocytes [10,15]. Myeloperoxidase is, however, absent from macrophages [15]. Serum/plasma levels of myeloperoxidase, therefore, probably reflect neutrophil turnover and activity [2,16]. A correlation between number of blood neutrophils and serum/plasma myeloperoxidase has also been found in the healthy [2]. The mode of catabolism of myeloperoxidase in the circulation is unknown. With the exception of serum lysozyme determinations, assays of lactoferrin and myeloperoxidase have not yet been established clinically. However, the potential usefulness of these proteins in the diagnosis and monitoring of macrophage/monocyte and neutrophil turnover and activity in various leukemias are becoming increasingly clear and other leucocyte-involved disorders [1,5-7,11,17-211. Since many of our patients are children, it has been necessary to establish reference ranges of the proteins in these age groups. To illustrate further the potential usefulness of serum determinations of the proteins, results from a group of infection-prone and leukemic children are also given. Material and methods Patients and reference groups There were three groups of children. One consisted girls) referred to our department for investigation
of 31 children of increased
(21 boys and 10 propensity for
I
11 7 6w 10 13 13 5 15 11 I 1
E.G. A.O. S.D. C.H. E.E. E.E. J.J. U.J. K.F. O.L. M.J.
Healthy children
Age
Patient
Serum levels of lactoferrin
TABLE
F F F F F F M M M M F
Sex
(LF), lysozyme
125-880
140 9 83 39 13700 390 570 230 298 40 164
children
70-410
1092 1232 911 40 8193 236 188 98 134 62 36
(Pdl)
MPO
(MPO) in leukemic
906-2 495
6004l 4513 61653 11700 23 200 3990 1442 861 1155 10 1183
LYS
(Pi0
LF
proteins
(Pm
Leucocyte
(LYS) and myeloperoxidase
AML (auer rods) AML AMML AMML CML CML ALL ALL ALL ALL Preleukemia (AML)
Diagnosis
Cytostatics 0 0 Cytostatics Cytostatics Cytostatics 0 0 0 Cytostatics 0
Therapy
E
124
bacterial infections (mean age, 5.1 years; range 1-14 years). Immunodeficiency had been excluded in all. All samples from these patients were taken at least 2 weeks after the last infection episode. Another group consisted of 10 children with diagnosed or suspected malignant diseases as presented in Table I. There were two children with acute myeloid leukemia (AML), two with acute myelo-monocytic leukemia (AMML), one with chronic myeloid leukemia (CML), four with acute lymphoblastic leukemia (ALL). One girl (M.J.), who for a year was suspected of having a preleukemic state, but who had normal peripheral blood and bone marrow at the time of investigation of leukocyte proteins, 2 months later developed a non-treatable acute myeloid leukemia (AML). Serum samples from the leukemic children were obtained during active stages of the disease except in patient E.E. in whom the second sample was obtained when the patient had a near normal blood cell count and morphology. All children except for J.J. and K.F. have died. The child reference group consisted of 31 children and were children referred to the hospital with diseases not supposed to affect leucocyte function, e.g. abdominal hernias. The age range in this group of children, 13 boys and 18 girls, was l-15 years (mean 8.9 years). The adult reference group consisted of 32 apparently healthy laboratory employees (15 males and 17 females) with an age range of 21-63 years. Methods Serum was separated from whole blood 60 min after the sample had been taken. Care was taken that coagulation occurred at room temperature and that it was complete before separation. Otherwise, spuriously high levels of leucocyte proteins may result. The sera were stored at - 70°C. PMNs were isolated from EDTA-blood by means of the Ficoll-Isopaque method of Boyurn [22] as previously described [16]. The purity of the PMNs was 96-100%. Extraction of the cells was carried out for 30 min at room temperature using 0.3% CTAB (cetyltrimethylammonium bromide) (BDH, Poole, UK) with 1 ml CTABsolution per 5 X lo6 PMNs. The cellular content of lactoferrin and myeloperoxidase was determined by means of radial immunodiffusion as already described [16]. Serum determinations of lactoferrin, lysozyme and myeloperoxidase were made by means of sensitive solid-phase radioimmunoassays [ 11,231. For statistical evaluation of differences between groups, Student’s t test was used after logarithmic transformation of the data. Results Serum levels of lactoferrin and myeloperoxidase (MPO) in the reference group of children are compared with the adult reference group in Fig. 1. The serum lactoferrin levels showed a tendency towards lower levels among children (geometric mean = 330 pg/l) than among adults (geometric mean = 395 pg/l). This is in contrast to the findings with serum MPO levels which, on average, were significantly (p = 0.01) higher (geometric mean = 174 pg/l) than the levels in adults (geometric mean = 121 pg/l). Serum lysozyme on the other hand did not differ significantly
125
P:ool
4-m . :. .
IOOO-
.:: .;.
2 3z 0
E
500-
.
.
2._
b
300-
5
200-
4:”
.
.:* .**
.!. l.
.** l.
A
.
I ADULTS
.I .*::.
2000.
2s...
.
l.
‘, _’ ‘.
l.:
IOO-
LOO=
::I*
:
d ;: 6 B ti
,$.l.: .:* .
22 .: :. .:* .:.. .
loo-
-7
. .
.. . .*. 7:r .:
l.
1.
I:
50-
. . .. *ii* :.._ .:* . ..* .. ..
.
; m
CHILDRENCHlLDREN
REFERENCES
P.0002
1r
600-
p
ADULTS
CHILDREN
REFERENCES
INFECTION-PRONE
CHILDREN INFECTION-PRONE
Fig. 1. Lactoferrin and myeloperoxidase in serum from normal and infection-prone children. The geometric means and the statistical evaluations of the differences between groups are indicated on the figures (n.s. = non-significant). The 95% confidence intervals for the respective proteins were for adults: MPO 40-370 pg/l, LF 190-810 pg/l; for children: MPO 70-410 pg/l, LF 125-880 pg/l; for infection-prone children: MPO 48-265 gg/l, LF 80-570 pg/l.
from the adult reference group (geometric means for children = 1638 pg/l and for adults = 1735 pg/l) (Fig. 2). In the infection-prone children, serum lactoferrin (geometric mean = 210 pgg/l) was significantly lower (p < 0.001) than in the child reference group (Fig. 1). Serum MPO (geometric mean = 113 pg/l) was also significantly lower (p = 0.002) in the infection-prone children when compared with the child reference group. Serum lysozyme was similar in the two groups of children (Fig. 2). Two possible causes of low serum levels of lactoferrin and MPO are a reduction of the intracellular content
ADULTS REFERENCES
CHILDREN
CHILDREN INFECTION
-PRONE
Fig. 2. Lysozyme in serum from normal and infection-prone children. The geometric means of the groups are indicated on the figure. There were no statistically significant differences between the three groups. The 95% confidence intervals for lysozyme in serum were for adults: 1060-2840 pg/l; for children: 906-2495 pg/l; for infection-prone children: 907-2460 pg/l.
126
of these proteins within the neutrophils and a reduction in the circulating cell number. In 15 of the infection-prone children (age range 2-14 years), intracellular quantitation of lactoferrin and MPO was performed (Fig. 3). The average content in this group of children was not significantly different from the adults. The average number of blood PMN among the reference children was 2.85 k 0.86 x 109/1 (n = 25) as compared with the number in the infection-prone group of children which on average was 3.73 f 1.58 x 109/1 (n = 17). This higher PMN number in the infection-prone group of children was statistically significant (p < 0.05). To exemplify the potential applicability of the measurement in serum of the three leucocyte-proteins lactoferrin, lysozyme and MPO, the results with 10 children with suspected or manifest leukemia are shown in Table I. In two patients (E.G. and A.O.) with AML, serum lactoferrin was below normal but the levels of MPO and lysozyme were raised. A similar pattern was seen in a patient with AMML (patient S.D.), but in addition, this patient had very high serum-lysozyme levels reflecting the monocytic involvement in the leukemia. Very high levels of serum lysozyme were observed in another patient with AMML (C.H.), but in this patient lactoferrin and MPO levels were both subnormal probably due to therapy. In one patient (E.E.) with CML, the levels of all proteins were very high. The lactoferrin and MPO levels became normal concomitant with a clinical and morphological return to normal. Serum lysozyme, however, remained at supranormal levels in spite of the normal morphology. In patients with ALL, the levels of the three proteins were unaffected in three cases (K.F., J.J. and U.J.), but in patient O.L. a profound reduction in lysozyme was observed following cytostatic therapy. In one patient (M.J.), with a preleukemia who developed AML 2 months after the blood-sampling, a subnormal level of serum MPO was observed with normal levels of the two other leucocyte proteins.
.
l:! -.1: :
LF
MPO
Fig. 3. Intracellular content of lactoferrin and myeloperoxidase in PMN from infection-prone children. The horizontal bars indicate the average content of the respective proteins. For comparison the intracellular content in PMNs obtained from 32 healthy adults are shown. These results are presented as means f 2 SD. There were no statistically significant differences between the two groups.
127
Discussion This paper has demonstrated the normal serum levels of lactoferrin, lysozyme and myeloperoxidase (MPO) in children. In the evaluation of such data it is important to consider the various mechanisms whereby these proteins will appear in serum and which factors determine the levels of the proteins at a given time. It is evident from the following that the interpretation must still be based on several assumptions. First of all, the cellular sources of the proteins have to be established. With respect to lactoferrin there is reason to believe that the neutrophil is the sole source contributing to the levels of this protein in blood [11,21]. Lysozyme is also contained within neutrophils, but probably the major part of lysozyme in serum originates from the secretion of monocytes and macrophages [S-7,11,20]. Whether in addition to the MPO from neutrophils, some also originates from monocytes, but not from macrophages, has not yet been clarified but seems probable [10,15]. Next, the mechanisms whereby these proteins are released from the cells to the extracellular environment has to be considered. This may be the result of (1) cell death, thus reflecting the turnover of cells in the circulation [l-4], or in the bone-marrow, (2) active release of proteins from circulating cells [l-4,16,24] or from cells participating in local processes [25], (3) active release from cells in vitro [24,26] or by accidentally damaged cells in vitro. Finally, the catabolism of the proteins has to be considered and, as pointed out in the introduction, nothing is known about myeloperoxidase. The elimination of lysozyme seems to be exclusively dependent on kidney function [8], whereas the elimination of lactoferrin is complex, probably implying the active participation of monocytes and macrophages [3,14]. The lactoferrin levels in normal children were similar to those in adults, but significantly lower in the infection-prone group of children in spite of an unaltered intracellular content and even higher cell counts in these children. MPO was also lower in the infection-prone children. Considering the above-mentioned conditions which determine the levels of the proteins in blood, explanations of these findings could be either reductions in the turnover and production of PMN in these children, or a reduced mobilization of the cells to the inflammatory sites. Indeed, a relationship between the serum lactoferrin levels and the chemotactic activity of serum has been observed recently, supporting the latter notion (unpublished Hakansson and Venge). Both mechanisms, however, should theoretically reduce the organism’s capacity to defend itself against invading organisms and the measurement of these proteins could, therefore, be of some importance in the evaluation of individuals with high infection-propensity. The significantly higher levels of MPO in spite of normal lactoferrin levels in normal children as compared with adults may have their biological counterpart in the bone marrow activity. In conditions such as pregnancy, high serum MPO levels have also been observed and been interpreted to signify an increased ineffective myelopoietic activity in the bone marrow [27]. In children, the bone marrow activity and the myelopoietic activity is considerably higher than in adults [28]. The high MPO levels could thus be a reflection of this fact. Assuming similar protein turnover rates in various leukemic disorders, the serum
128
level of a leucocyte-derived protein probably reflects the size of the cell population at a given time capable of synthesizing the respective proteins. In normal myelopoiesis the MPO-containing, primary (azurophil) granules are formed in the promyelocytic [10,29], in contrast to the secondary (specific), lactoferrin-containing granules which are formed in the myelocytic stage [10,29]. In patients with acute myeloid leukemia, the serum levels of these two proteins could, therefore, provide an estimate of the stage of maturation and the size of the leukemic cell mass. An almost normal level of serum lactoferrin in one patient with acute myeloid leukemia (E.G.) suggested that a substantial number of cells had passed the myelocytic stage. The other patient with AML had almost undetectable lactoferrin levels suggesting an arrest in the maturation before the myelocytic stage. The very high serum MPO levels in both demonstrated that there was an accumulation of cells arrested at the promyelocytic stage. In adult patients with AML, high and normal serum levels of lysozyme [30] and MPO [21] have been described as the most common findings and interpreted as reflecting blastic synthesis of the proteins or possibly ineffective granulocytopoiesis in the bone marrow. Conversely, normal or low levels of serum lactoferrin were the most common finding in adult AML [21], probably due to the scarcity of mature cells in the granulopoiesis or the low intracellular content within the morphologically mature neutrophil [21,31]. In the cases with AML, the serum lysozyme may have been derived either from myeloid cells or from monocytic cells. When serum lysozyme levels were as high as in patients S.D. and C.H., the origin of lysozyme was most likely to be the monocyte [7]. Equally high levels of lysozyme to those seen in myelo-monocytic leukemia may also be seen in patients with chronic myelocytic leukemia ([l] and unpublished observation). In CML, however, serum lactoferrin and serum MPO are expected to follow the serum lysozyme levels, as is seen in patient E.E., since no maturation block is expected in CML not complicated by a blastic crisis [19]. In patients with ALL, serum determinations of the three neutrophil and monocyte/macrophage derived proteins were not expected to offer much information as to the malignancy per se. Additional information as to the condition of monocytopoiesis and myelopoiesis may, however, be obtained since apparent monocytopenia and neutropenia are common findings in ALL (281. The profound reduction in serum lysozyme in patient O.L. with ALL was particularly interesting since this could indicate a complete absence of lysozyme synthezising monocytes/macrophages but, as indicated by the lactoferrin and MPO levels, the presence of some myelopoiesis. The low serum MPO level in the patient with preleukemia may suggest a disturbance of the cell maturation not yet evident from the morphology of the peripheral blood cells. Although a limited number of children with hematological malignancies have been studied, the combined serum determinations of lactoferrin, lysozyme and myeloperoxidase seem to be a valuable complement to the morphological diagnosis of these diseases and may also provide a future basis for the monitoring during therapy. The utilisation of these proteins in the evaluation of cell-production and turnover could also provide new insight into mechanisms of increased infection propensity. It is, however, apparent from the present study that proper reference ranges have to be established to be successful in these attempts.
129
Acknowledgements The technical assistance of Ms. Lena Bjiirkman and Ms. Kerstin Lindblad is greatly appreciated. This work was supported by the Swedish Medical Research Council and the medical faculty of the University of Uppsala. References 1 Hansen NE, Plasma lysozyme a measure of neutrophil turnover. An analytical review. Ser Haematol 1974; VII: l-87. 2 Hansen NE, Malmquist J, Thorell J. Plasma myeloperoxidase and lactoferrin measured by radioimmunoassay: Relations to neutrophil kinetics. Acta Med Stand 1975; 198: 437-443. 3 Bennett RM, Kokocinski T. Lactoferrin turnover in man. Clin Sci 1979; 57: 453-460. 4 Hansen NE. The relationship between the turnover rate of neutrophilic granulocytes and plasma lysozyme levels. Br J Haematol 1973; 25: 771-782. 5 Pascual RS, Gee JBL, Finch SC. Usefulness of serum lysozyme measurement in diagnosis and evaluation of sarcoidosis. N Eng J Med 1973; 289: 1074-1076. 6 Falchuk KR, Perrotto JL, Isselbacher KJ. Serum lysozyme in Crohn’s disease and ulcerative colitis. N Eng J Med 1975; 299: 395-397. 7 Osserman EF, Lawlor DP. Serum and urinary lysozyme (muramidase) in monocytic and monomyelocytic leukemia. J Exp Med 1966; 124: 921-952. 8 Hansen NE, Karle H, Andersen V, Olgaard K. Lysozyme turnover in man. J Clin Invest 1972; 51: 1146-1155. 9 Johansson BG, Ravnskov U. The serum level and urinary excretion of a,-microglobulin, &-microglobulin and lysozyme in renal disease. Stand J Urol Nephrol 1972; 6: 249-256. 10 Spitznagel JK, Dalldorf FG, Leffell MS, Folds JD, Welsh JRH, Cooney BS, Martin BS. Character of azurophil and specific granules purified from human polymorphonuclear leucocytes. Lab Invest 1974; 30: 774-7x5. 11 Olofsson T, Olsson I, Venge P, Elgefors B. Serum myeloperoxidase and lactoferrin in neutropenia. Stand J Haematol 1977; 18: 73-80. 12 Bennet RM, Mohla C. A solid-phase radioimmunoassay for the measurement of lactoferrin in human plasma: Variations with age, sex and disease. J Lab Clin Med 1976; 88: 156-166. 13 Bennett RM, Kokocinski T. Lactoferrin content of peripheral blood cells. Br J Haematol 1978; 39: 509-521. 14 Van Snick JL, Masson PL. The binding of human lactoferrin to mouse peritoneal cells. J Exp Med 1976; 144: 1568-1580. 15 Nichols BA, Bainton DF. Differentiation of human monocytes in bone marrow and blood. Sequential formation of two granule populations. Lab Invest 1973; 29: 27-40. 16 Venge P, Stramberg A, Braconier JH, Roxin LE, Olsson I. Neutrophil and eosinophil granulocytes in bacterial infection: Sequential studies of cellular and serum levels of granule proteins. Br J Haematol 1978; 38: 475-483. 17 Malmquist J. Serum l.actoferrin in leukemia and polycythaemia Vera. Stand J Haematol 1972; 9: 305-310. 18 Malmquist J. Serum myeloperoxidase in leukemia and polycythaemia Vera. Stand J Haematol 1972; 9: 311-317. 19 Olofsson T, Olsson I, Venge P. Myeloperoxidase and lactoferrin of blood neutrophils and plasma in chronic granulocytic leukemia. Stand J Haematol 1977; 18: 113-120. 20 Hansen NE, Karle H. Elevated plasma lysozyme in Hodgkin’s disease. Stand .I Haematol 1979; 22: 173-178. 21 Olsson I, Olofsson T, Ohlsson K, Gustavsson A. Serum and plasma myeloperoxidase, elastase and lactoferrin content in acute myeloid leukemia. Stand J Haematol 1979; 22: 397-406.
130 22 Boyum A. Separation of leucocytes from blood and bone marrow. Stand J Chn Lab Invest 1968; 21, Suppl. 91: 77-89. 23 Venge P, HaBgren R, Stdenheim G, Olsson I. Effects of serum and cations on the selective release of granular proteins from human neutrophils during phagocytosis. Stand J Haematol. 1979; 22: 317-326. 24 Venge P, Boberg M, H&kansson L, Peterson C. Identification of a serum-derived promotor of granulocyte granule secretion: Study on a patient with chronic pruritus. Immunology 1982; 45: 521-529. 25 Hal&en R, Venge P, Wikstrbm B. Hemodialysis-induced increase in serum-lactoferrin and serum eosinophil cationic protein as signs of local neutrophil and eosinophil degranulation. Nephron 1981; 29: 233-238. 26 Plow EF. Leukocyte elastase release during blood coagulation. A potential mechanism for activation of the alternative fibrinolytic pathway. J Clin Invest 1982; 69: 564-572. 27 ijberg G, Lindmark G, Moberg L, Venge P. Serum and cellular levels of lactoferrin, lysozyme and myeloperoxidase during pregnancy. In preparation. 28 Miale JB. In: Laboratory medicine. Hematology. St. Louis, USA: The C.V. Mosby Company, Fifth edition, 1977. 29 Bainton DF, Ullyot JL, Farquhar UC. The development of neutrophilic polymorphonuclear leucocytes in human bone marrow. Origin and content of azurophil and specific granules. J Exp Med 1971; 134: 907-934. 30 Karle H, Hansen NE, Killman SA. Intracellular lysozyme in mature neutrophils and blast cells in acute leukemia. Blood 1974; 44: 247-255. 31 Odeberg H, Olofsson T, Olsson I. Primary and secondary granule contents and bactericidal capacity of neutrophils in acute leukemia. Blood Cells 1976; 2: 519-551.
Clinica Chimica Acta, 136 (1984) 121-130 Elsevier
CCA 02734
Serum-levels of lactoferrin, lysazyme and myeloperoxidase in normal, infection-prone and leukemic children Per Venge a, Tony Foucard Departments
b, J&n Henriksen Anders Kreuger b
‘, Lena
H&ansson
a and
of” Clinical Chemistry and b Pediatrics, University Hospital, Uppsala (Sweden) and ’ Department o/Pediatrics, University Hospital, A’rhus (Denmark) (Received April 15th; revision August 18th, 1983) Key words: Blood ceil; Turnover; Granule prctein; Radioimmunoassay
Serum levels of lactoferrin, lysozyme and myeloperoxidase have been established in 31 healthy children. On average, serum lactoferrin was 330 pg/l, serum lysozyme 1638 pg/l and serum myeloperoxidase 174 pg/l. Serum myeloperoxidase was, on average, significantly higher in children than in adults (p = O.Ol), whereas serum lactoferrin and serum lysozyme were equal to those of adults. In a group of infection-prone children (n = 31), both serum lactoferrin and serum myeloperoxidase, but not the serum lysozyme levels, were significantly lower (p < 0.001 and p = 0.002, respectively) than those of the reference children in spite of normal intracellular contents and even somewhat higher peripheral blood polymorphonuclear counts. Based on the assumption that serum lactoferrin and serum myeloperoxidase reflect turnover and activity of neutrophil granulocytes, the findings could suggest reduction in these respects and could be one contributing factor to the high infection propensity of these children. Serum levels of .the three proteins have also been measured in 10 children with suspected or various forms of manifest leukemia. It is suggested that the levels reflect turnover and stage of maturation of the myeloid and monocytic cells and could, therefore, aid in the understanding and diagnosis of these diseases.
Correspondence and reprint requests to: Per Venge, MD, Department of Clinical Chemistry, University Hospital, S-750 14 Uppsala, Sweden. 0009-8981/84/$03.00
0 1984 Elsevier Science Publishers B.V.
122
Introduction Turnover and activity of leucocytes may be estimated by serum or plasma determination of leucocyte-derived proteins [l-3]. Lysozyme has been most extensively utilized for this purpose and in serum is probably derived from both neutrophils and monocytes/macrophages [1,4]. The relative contribution from these two sources is, however, uncertain. In the healthy subjects, a correlation with the blood neutrophil count is obtained [4], but in chronic inflammatory diseases such as sarcoidosis or Crohn’s disease [5,6], the high serum lysozyme levels most likely reflect monocyte/macrophage activity. In monocytic leukemia, serum lysozyme determinations are used to establish the diagnosis [7] and to estimate the leukemic cell mass [l]. Lysozyme is eliminated from the circulation through glomerular filtration (GFR) [8] and serum lysozyme levels are consequently raised when there is a reduction in GFR [8,9]. Lactoferrin is an iron-binding protein contained within the secondary (specific) granules of neutrophils [lo] and serum/plasma levels of this protein are probably specific markers of neutrophil turnover and activity [2,3,11]. Serum/plasma levels of lactoferrin in the healthy correlate with the number of blood neutrophils [2]. One report [12] has shown a 50% reduction in the plasma levels in women under the age of 50 years, probably due to the reduction of lactoferrin within the circulating neutrophil [13]. Monocytes/macrophages seem to have specific receptors for lactoferrin [14]. The catabolism may, therefore, be dependent on the availability of these structures. An uptake of lactoferrin in the liver also points to a dependence on liver function for the elimination of lactoferrin [3]. Myeloperoxidase is contained within the primary (azurophil) granules of neutrophils but also to some extent within the monocytes [10,15]. Myeloperoxidase is, however, absent from macrophages [15]. Serum/plasma levels of myeloperoxidase, therefore, probably reflect neutrophil turnover and activity [2,16]. A correlation between number of blood neutrophils and serum/plasma myeloperoxidase has also been found in the healthy [2]. The mode of catabolism of myeloperoxidase in the circulation is unknown. With the exception of serum lysozyme determinations, assays of lactoferrin and myeloperoxidase have not yet been established clinically. However, the potential usefulness of these proteins in the diagnosis and monitoring of macrophage/monocyte and neutrophil turnover and activity in various leukemias are becoming increasingly clear and other leucocyte-involved disorders [1,5-7,11,17-211. Since many of our patients are children, it has been necessary to establish reference ranges of the proteins in these age groups. To illustrate further the potential usefulness of serum determinations of the proteins, results from a group of infection-prone and leukemic children are also given. Material and methods Patients and reference groups There were three groups of children. One consisted girls) referred to our department for investigation
of 31 children of increased
(21 boys and 10 propensity for
I
11 7 6w 10 13 13 5 15 11 I 1
E.G. A.O. S.D. C.H. E.E. E.E. J.J. U.J. K.F. O.L. M.J.
Healthy children
Age
Patient
Serum levels of lactoferrin
TABLE
F F F F F F M M M M F
Sex
(LF), lysozyme
125-880
140 9 83 39 13700 390 570 230 298 40 164
children
70-410
1092 1232 911 40 8193 236 188 98 134 62 36
(Pdl)
MPO
(MPO) in leukemic
906-2 495
6004l 4513 61653 11700 23 200 3990 1442 861 1155 10 1183
LYS
(Pi0
LF
proteins
(Pm
Leucocyte
(LYS) and myeloperoxidase
AML (auer rods) AML AMML AMML CML CML ALL ALL ALL ALL Preleukemia (AML)
Diagnosis
Cytostatics 0 0 Cytostatics Cytostatics Cytostatics 0 0 0 Cytostatics 0
Therapy
E
124
bacterial infections (mean age, 5.1 years; range 1-14 years). Immunodeficiency had been excluded in all. All samples from these patients were taken at least 2 weeks after the last infection episode. Another group consisted of 10 children with diagnosed or suspected malignant diseases as presented in Table I. There were two children with acute myeloid leukemia (AML), two with acute myelo-monocytic leukemia (AMML), one with chronic myeloid leukemia (CML), four with acute lymphoblastic leukemia (ALL). One girl (M.J.), who for a year was suspected of having a preleukemic state, but who had normal peripheral blood and bone marrow at the time of investigation of leukocyte proteins, 2 months later developed a non-treatable acute myeloid leukemia (AML). Serum samples from the leukemic children were obtained during active stages of the disease except in patient E.E. in whom the second sample was obtained when the patient had a near normal blood cell count and morphology. All children except for J.J. and K.F. have died. The child reference group consisted of 31 children and were children referred to the hospital with diseases not supposed to affect leucocyte function, e.g. abdominal hernias. The age range in this group of children, 13 boys and 18 girls, was l-15 years (mean 8.9 years). The adult reference group consisted of 32 apparently healthy laboratory employees (15 males and 17 females) with an age range of 21-63 years. Methods Serum was separated from whole blood 60 min after the sample had been taken. Care was taken that coagulation occurred at room temperature and that it was complete before separation. Otherwise, spuriously high levels of leucocyte proteins may result. The sera were stored at - 70°C. PMNs were isolated from EDTA-blood by means of the Ficoll-Isopaque method of Boyurn [22] as previously described [16]. The purity of the PMNs was 96-100%. Extraction of the cells was carried out for 30 min at room temperature using 0.3% CTAB (cetyltrimethylammonium bromide) (BDH, Poole, UK) with 1 ml CTABsolution per 5 X lo6 PMNs. The cellular content of lactoferrin and myeloperoxidase was determined by means of radial immunodiffusion as already described [16]. Serum determinations of lactoferrin, lysozyme and myeloperoxidase were made by means of sensitive solid-phase radioimmunoassays [ 11,231. For statistical evaluation of differences between groups, Student’s t test was used after logarithmic transformation of the data. Results Serum levels of lactoferrin and myeloperoxidase (MPO) in the reference group of children are compared with the adult reference group in Fig. 1. The serum lactoferrin levels showed a tendency towards lower levels among children (geometric mean = 330 pg/l) than among adults (geometric mean = 395 pg/l). This is in contrast to the findings with serum MPO levels which, on average, were significantly (p = 0.01) higher (geometric mean = 174 pg/l) than the levels in adults (geometric mean = 121 pg/l). Serum lysozyme on the other hand did not differ significantly
125
P:ool
4-m . :. .
IOOO-
.:: .;.
2 3z 0
E
500-
.
.
2._
b
300-
5
200-
4:”
.
.:* .**
.!. l.
.** l.
A
.
I ADULTS
.I .*::.
2000.
2s...
.
l.
‘, _’ ‘.
l.:
IOO-
LOO=
::I*
:
d ;: 6 B ti
,$.l.: .:* .
22 .: :. .:* .:.. .
loo-
-7
. .
.. . .*. 7:r .:
l.
1.
I:
50-
. . .. *ii* :.._ .:* . ..* .. ..
.
; m
CHILDRENCHlLDREN
REFERENCES
P.0002
1r
600-
p
ADULTS
CHILDREN
REFERENCES
INFECTION-PRONE
CHILDREN INFECTION-PRONE
Fig. 1. Lactoferrin and myeloperoxidase in serum from normal and infection-prone children. The geometric means and the statistical evaluations of the differences between groups are indicated on the figures (n.s. = non-significant). The 95% confidence intervals for the respective proteins were for adults: MPO 40-370 pg/l, LF 190-810 pg/l; for children: MPO 70-410 pg/l, LF 125-880 pg/l; for infection-prone children: MPO 48-265 gg/l, LF 80-570 pg/l.
from the adult reference group (geometric means for children = 1638 pg/l and for adults = 1735 pg/l) (Fig. 2). In the infection-prone children, serum lactoferrin (geometric mean = 210 pgg/l) was significantly lower (p < 0.001) than in the child reference group (Fig. 1). Serum MPO (geometric mean = 113 pg/l) was also significantly lower (p = 0.002) in the infection-prone children when compared with the child reference group. Serum lysozyme was similar in the two groups of children (Fig. 2). Two possible causes of low serum levels of lactoferrin and MPO are a reduction of the intracellular content
ADULTS REFERENCES
CHILDREN
CHILDREN INFECTION
-PRONE
Fig. 2. Lysozyme in serum from normal and infection-prone children. The geometric means of the groups are indicated on the figure. There were no statistically significant differences between the three groups. The 95% confidence intervals for lysozyme in serum were for adults: 1060-2840 pg/l; for children: 906-2495 pg/l; for infection-prone children: 907-2460 pg/l.
126
of these proteins within the neutrophils and a reduction in the circulating cell number. In 15 of the infection-prone children (age range 2-14 years), intracellular quantitation of lactoferrin and MPO was performed (Fig. 3). The average content in this group of children was not significantly different from the adults. The average number of blood PMN among the reference children was 2.85 k 0.86 x 109/1 (n = 25) as compared with the number in the infection-prone group of children which on average was 3.73 f 1.58 x 109/1 (n = 17). This higher PMN number in the infection-prone group of children was statistically significant (p < 0.05). To exemplify the potential applicability of the measurement in serum of the three leucocyte-proteins lactoferrin, lysozyme and MPO, the results with 10 children with suspected or manifest leukemia are shown in Table I. In two patients (E.G. and A.O.) with AML, serum lactoferrin was below normal but the levels of MPO and lysozyme were raised. A similar pattern was seen in a patient with AMML (patient S.D.), but in addition, this patient had very high serum-lysozyme levels reflecting the monocytic involvement in the leukemia. Very high levels of serum lysozyme were observed in another patient with AMML (C.H.), but in this patient lactoferrin and MPO levels were both subnormal probably due to therapy. In one patient (E.E.) with CML, the levels of all proteins were very high. The lactoferrin and MPO levels became normal concomitant with a clinical and morphological return to normal. Serum lysozyme, however, remained at supranormal levels in spite of the normal morphology. In patients with ALL, the levels of the three proteins were unaffected in three cases (K.F., J.J. and U.J.), but in patient O.L. a profound reduction in lysozyme was observed following cytostatic therapy. In one patient (M.J.), with a preleukemia who developed AML 2 months after the blood-sampling, a subnormal level of serum MPO was observed with normal levels of the two other leucocyte proteins.
.
l:! -.1: :
LF
MPO
Fig. 3. Intracellular content of lactoferrin and myeloperoxidase in PMN from infection-prone children. The horizontal bars indicate the average content of the respective proteins. For comparison the intracellular content in PMNs obtained from 32 healthy adults are shown. These results are presented as means f 2 SD. There were no statistically significant differences between the two groups.
127
Discussion This paper has demonstrated the normal serum levels of lactoferrin, lysozyme and myeloperoxidase (MPO) in children. In the evaluation of such data it is important to consider the various mechanisms whereby these proteins will appear in serum and which factors determine the levels of the proteins at a given time. It is evident from the following that the interpretation must still be based on several assumptions. First of all, the cellular sources of the proteins have to be established. With respect to lactoferrin there is reason to believe that the neutrophil is the sole source contributing to the levels of this protein in blood [11,21]. Lysozyme is also contained within neutrophils, but probably the major part of lysozyme in serum originates from the secretion of monocytes and macrophages [S-7,11,20]. Whether in addition to the MPO from neutrophils, some also originates from monocytes, but not from macrophages, has not yet been clarified but seems probable [10,15]. Next, the mechanisms whereby these proteins are released from the cells to the extracellular environment has to be considered. This may be the result of (1) cell death, thus reflecting the turnover of cells in the circulation [l-4], or in the bone-marrow, (2) active release of proteins from circulating cells [l-4,16,24] or from cells participating in local processes [25], (3) active release from cells in vitro [24,26] or by accidentally damaged cells in vitro. Finally, the catabolism of the proteins has to be considered and, as pointed out in the introduction, nothing is known about myeloperoxidase. The elimination of lysozyme seems to be exclusively dependent on kidney function [8], whereas the elimination of lactoferrin is complex, probably implying the active participation of monocytes and macrophages [3,14]. The lactoferrin levels in normal children were similar to those in adults, but significantly lower in the infection-prone group of children in spite of an unaltered intracellular content and even higher cell counts in these children. MPO was also lower in the infection-prone children. Considering the above-mentioned conditions which determine the levels of the proteins in blood, explanations of these findings could be either reductions in the turnover and production of PMN in these children, or a reduced mobilization of the cells to the inflammatory sites. Indeed, a relationship between the serum lactoferrin levels and the chemotactic activity of serum has been observed recently, supporting the latter notion (unpublished Hakansson and Venge). Both mechanisms, however, should theoretically reduce the organism’s capacity to defend itself against invading organisms and the measurement of these proteins could, therefore, be of some importance in the evaluation of individuals with high infection-propensity. The significantly higher levels of MPO in spite of normal lactoferrin levels in normal children as compared with adults may have their biological counterpart in the bone marrow activity. In conditions such as pregnancy, high serum MPO levels have also been observed and been interpreted to signify an increased ineffective myelopoietic activity in the bone marrow [27]. In children, the bone marrow activity and the myelopoietic activity is considerably higher than in adults [28]. The high MPO levels could thus be a reflection of this fact. Assuming similar protein turnover rates in various leukemic disorders, the serum
128
level of a leucocyte-derived protein probably reflects the size of the cell population at a given time capable of synthesizing the respective proteins. In normal myelopoiesis the MPO-containing, primary (azurophil) granules are formed in the promyelocytic [10,29], in contrast to the secondary (specific), lactoferrin-containing granules which are formed in the myelocytic stage [10,29]. In patients with acute myeloid leukemia, the serum levels of these two proteins could, therefore, provide an estimate of the stage of maturation and the size of the leukemic cell mass. An almost normal level of serum lactoferrin in one patient with acute myeloid leukemia (E.G.) suggested that a substantial number of cells had passed the myelocytic stage. The other patient with AML had almost undetectable lactoferrin levels suggesting an arrest in the maturation before the myelocytic stage. The very high serum MPO levels in both demonstrated that there was an accumulation of cells arrested at the promyelocytic stage. In adult patients with AML, high and normal serum levels of lysozyme [30] and MPO [21] have been described as the most common findings and interpreted as reflecting blastic synthesis of the proteins or possibly ineffective granulocytopoiesis in the bone marrow. Conversely, normal or low levels of serum lactoferrin were the most common finding in adult AML [21], probably due to the scarcity of mature cells in the granulopoiesis or the low intracellular content within the morphologically mature neutrophil [21,31]. In the cases with AML, the serum lysozyme may have been derived either from myeloid cells or from monocytic cells. When serum lysozyme levels were as high as in patients S.D. and C.H., the origin of lysozyme was most likely to be the monocyte [7]. Equally high levels of lysozyme to those seen in myelo-monocytic leukemia may also be seen in patients with chronic myelocytic leukemia ([l] and unpublished observation). In CML, however, serum lactoferrin and serum MPO are expected to follow the serum lysozyme levels, as is seen in patient E.E., since no maturation block is expected in CML not complicated by a blastic crisis [19]. In patients with ALL, serum determinations of the three neutrophil and monocyte/macrophage derived proteins were not expected to offer much information as to the malignancy per se. Additional information as to the condition of monocytopoiesis and myelopoiesis may, however, be obtained since apparent monocytopenia and neutropenia are common findings in ALL (281. The profound reduction in serum lysozyme in patient O.L. with ALL was particularly interesting since this could indicate a complete absence of lysozyme synthezising monocytes/macrophages but, as indicated by the lactoferrin and MPO levels, the presence of some myelopoiesis. The low serum MPO level in the patient with preleukemia may suggest a disturbance of the cell maturation not yet evident from the morphology of the peripheral blood cells. Although a limited number of children with hematological malignancies have been studied, the combined serum determinations of lactoferrin, lysozyme and myeloperoxidase seem to be a valuable complement to the morphological diagnosis of these diseases and may also provide a future basis for the monitoring during therapy. The utilisation of these proteins in the evaluation of cell-production and turnover could also provide new insight into mechanisms of increased infection propensity. It is, however, apparent from the present study that proper reference ranges have to be established to be successful in these attempts.
129
Acknowledgements The technical assistance of Ms. Lena Bjiirkman and Ms. Kerstin Lindblad is greatly appreciated. This work was supported by the Swedish Medical Research Council and the medical faculty of the University of Uppsala. References 1 Hansen NE, Plasma lysozyme a measure of neutrophil turnover. An analytical review. Ser Haematol 1974; VII: l-87. 2 Hansen NE, Malmquist J, Thorell J. Plasma myeloperoxidase and lactoferrin measured by radioimmunoassay: Relations to neutrophil kinetics. Acta Med Stand 1975; 198: 437-443. 3 Bennett RM, Kokocinski T. Lactoferrin turnover in man. Clin Sci 1979; 57: 453-460. 4 Hansen NE. The relationship between the turnover rate of neutrophilic granulocytes and plasma lysozyme levels. Br J Haematol 1973; 25: 771-782. 5 Pascual RS, Gee JBL, Finch SC. Usefulness of serum lysozyme measurement in diagnosis and evaluation of sarcoidosis. N Eng J Med 1973; 289: 1074-1076. 6 Falchuk KR, Perrotto JL, Isselbacher KJ. Serum lysozyme in Crohn’s disease and ulcerative colitis. N Eng J Med 1975; 299: 395-397. 7 Osserman EF, Lawlor DP. Serum and urinary lysozyme (muramidase) in monocytic and monomyelocytic leukemia. J Exp Med 1966; 124: 921-952. 8 Hansen NE, Karle H, Andersen V, Olgaard K. Lysozyme turnover in man. J Clin Invest 1972; 51: 1146-1155. 9 Johansson BG, Ravnskov U. The serum level and urinary excretion of a,-microglobulin, &-microglobulin and lysozyme in renal disease. Stand J Urol Nephrol 1972; 6: 249-256. 10 Spitznagel JK, Dalldorf FG, Leffell MS, Folds JD, Welsh JRH, Cooney BS, Martin BS. Character of azurophil and specific granules purified from human polymorphonuclear leucocytes. Lab Invest 1974; 30: 774-7x5. 11 Olofsson T, Olsson I, Venge P, Elgefors B. Serum myeloperoxidase and lactoferrin in neutropenia. Stand J Haematol 1977; 18: 73-80. 12 Bennet RM, Mohla C. A solid-phase radioimmunoassay for the measurement of lactoferrin in human plasma: Variations with age, sex and disease. J Lab Clin Med 1976; 88: 156-166. 13 Bennett RM, Kokocinski T. Lactoferrin content of peripheral blood cells. Br J Haematol 1978; 39: 509-521. 14 Van Snick JL, Masson PL. The binding of human lactoferrin to mouse peritoneal cells. J Exp Med 1976; 144: 1568-1580. 15 Nichols BA, Bainton DF. Differentiation of human monocytes in bone marrow and blood. Sequential formation of two granule populations. Lab Invest 1973; 29: 27-40. 16 Venge P, Stramberg A, Braconier JH, Roxin LE, Olsson I. Neutrophil and eosinophil granulocytes in bacterial infection: Sequential studies of cellular and serum levels of granule proteins. Br J Haematol 1978; 38: 475-483. 17 Malmquist J. Serum l.actoferrin in leukemia and polycythaemia Vera. Stand J Haematol 1972; 9: 305-310. 18 Malmquist J. Serum myeloperoxidase in leukemia and polycythaemia Vera. Stand J Haematol 1972; 9: 311-317. 19 Olofsson T, Olsson I, Venge P. Myeloperoxidase and lactoferrin of blood neutrophils and plasma in chronic granulocytic leukemia. Stand J Haematol 1977; 18: 113-120. 20 Hansen NE, Karle H. Elevated plasma lysozyme in Hodgkin’s disease. Stand .I Haematol 1979; 22: 173-178. 21 Olsson I, Olofsson T, Ohlsson K, Gustavsson A. Serum and plasma myeloperoxidase, elastase and lactoferrin content in acute myeloid leukemia. Stand J Haematol 1979; 22: 397-406.
130 22 Boyum A. Separation of leucocytes from blood and bone marrow. Stand J Chn Lab Invest 1968; 21, Suppl. 91: 77-89. 23 Venge P, HaBgren R, Stdenheim G, Olsson I. Effects of serum and cations on the selective release of granular proteins from human neutrophils during phagocytosis. Stand J Haematol. 1979; 22: 317-326. 24 Venge P, Boberg M, H&kansson L, Peterson C. Identification of a serum-derived promotor of granulocyte granule secretion: Study on a patient with chronic pruritus. Immunology 1982; 45: 521-529. 25 Hal&en R, Venge P, Wikstrbm B. Hemodialysis-induced increase in serum-lactoferrin and serum eosinophil cationic protein as signs of local neutrophil and eosinophil degranulation. Nephron 1981; 29: 233-238. 26 Plow EF. Leukocyte elastase release during blood coagulation. A potential mechanism for activation of the alternative fibrinolytic pathway. J Clin Invest 1982; 69: 564-572. 27 ijberg G, Lindmark G, Moberg L, Venge P. Serum and cellular levels of lactoferrin, lysozyme and myeloperoxidase during pregnancy. In preparation. 28 Miale JB. In: Laboratory medicine. Hematology. St. Louis, USA: The C.V. Mosby Company, Fifth edition, 1977. 29 Bainton DF, Ullyot JL, Farquhar UC. The development of neutrophilic polymorphonuclear leucocytes in human bone marrow. Origin and content of azurophil and specific granules. J Exp Med 1971; 134: 907-934. 30 Karle H, Hansen NE, Killman SA. Intracellular lysozyme in mature neutrophils and blast cells in acute leukemia. Blood 1974; 44: 247-255. 31 Odeberg H, Olofsson T, Olsson I. Primary and secondary granule contents and bactericidal capacity of neutrophils in acute leukemia. Blood Cells 1976; 2: 519-551.