| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
1
Department of Clinical Chemistry, Royal Liverpool University Hospital, Prescot Street, Liverpool L7 8XP, United Kingdom
2
Countess of Chester Hospital, Liverpool Road, Chester, United Kingdom
3
Southend Hospital, Westcliffe-on-Sea, Essex SS0 0RY, United Kingdom
a author for correspondence: fax 44-151-706-5813, e-mail mjdiver{at}liv.ac.uk
 |
Introduction |
|---|
 Top Introduction References |
|---|
We report the case of a 30-year-old woman who initially attended her primary care physician because of the onset of painful irregular periods. Her cycle usually was regular, but she had had an 8-week interval of amenorrhea, followed by a particularly painful bleed for which she sought medical advice. Before this, and subsequently, her menstruation had been completely regular with a 28-day cycle. She had no other problems.
The patient’s initial serum prolactin was recorded as 15 800
mIU/L (
530 µg/L) in a Bayer Immuno
1TM assay (Bayer Corporation). Other
investigations at the time were entirely normal.
When the subject was monitored 2 months later, she was
symptomless and menstruating regularly; her serum prolactin, using the
same assay as before, was 8440 mIU/L (270 µg/L). Pituitary imaging
by magnetic resonance was normal. She had, of choice, never been
pregnant.
Because of the patient’s lack of symptoms, normal pituitary imaging, and regular cycles, further analytical investigations were carried out on a sample of her serum. After polyethylene glycol (PEG) precipitation, the recovery of prolactin was low, indicating the presence of macroprolactin (2), and 15% of her total prolactin was estimated to be monomeric prolactin.
The patient’s serum prolactin concentration was remeasured using a
Wallac DelfiaTM assay (EG & G Wallac) and
compared with that measured previously on the Bayer Immuno 1
analyzer. The Delfia assay recorded a prolactin concentration of 2980
mIU/L (100 µg/L), a difference of 70%. This marked discrepancy
between the results prompted measurement of prolactin in the specimen
by three other commonly used automated immunoanalyzers (Table 1
).
|
In our experience, the Delfia, Immuno 1, and Elecsys (Roche Diagnostics) assays are consistently high-reacting assays with samples containing macroprolactin, and the ACS:180 (Bayer) is, most frequently, a relatively low-reacting assay (3). In 24 cases of macroprolactinemia, results from the Delfia assay always exceeded those from the ACS:180 assay (4), presumably because the IgG antibody masks the epitope of prolactin with which the ACS:180 prolactin antibody reacts.
The presence of macroprolactinemia was confirmed by the Delfia assay
and further investigated by the following techniques (see Table 1
for
results): (a) PEG precipitation (2);
(b) 125I binding (5);
(c) ultrafiltration (6); (d) gel
filtration chromatography (2); and (e) protein G
affinity chromatography (5).
Treatment of normal serum containing monomeric prolactin with 4 mol/L
urea causes some denaturation of immunoreactive prolactin detected by
both the Delfia and ACS:180 assays. In our experience, treatment of
serum containing macroprolactin converts this to monomeric prolactin
(Fig. 1A
), decreasing the amount of immunoreactive prolactin detected
by the Delfia assay and increasing the amount detected by the ACS:180
assay.
|
Serum from our patient behaved differently, with an increase in
immunoreactive prolactin in the Delfia assay and a decrease with the
ACS:180 after treatment with urea as well as a more complex pattern on
gel filtration chromatography (Fig. 1B
).
To further examine the glycoprotein content of this novel high-molecular mass form of prolactin, we analyzed serum samples after lectin affinity chromatography. Using the immobilized lectin concanavalin A linked to SepharoseTM (5), we found that 36% of the immunoreactive prolactin from our patient’s serum was retained on the concanavalin A-Sepharose column compared with <10% from samples containing only simple prolactin. The macroprolactin peak accounted for 91% of the total immunoreactive prolactin in the Delfia assay. Using the peaks of albumin and monomeric prolactin to define the relationship between elution volume and molecular mass, we estimated that the molecular mass of the macroprolactin from this patient was 234 kDa. In 13 other cases of macroprolactin, the mean molecular mass was estimated at 162 kDa (range, 119–237 kDa).
After these original investigations, the patient remained well, with
continued regular periods, on no therapy, and with no other problems.
After a 2-year interval, she became pregnant, and a serum prolactin
measured at 6 months of gestation revealed (Bayer Immuno 1) a
concentration of 29 949 mIU/L (1000 µg/L), of which 1700 mIU/L
(5.6%) was found to be monomeric after PEG precipitation. The
pregnancy was uneventful, and the patient was delivered of a 3.6-kg boy
following induction at 42 weeks of gestation. She lactated without
difficulty, but decided not to breastfeed the baby.
Two weeks postpartum, her serum prolactin measured on the Bayer Immuno
1 analyzer was 42 000 mIU/L (1400 µg/L), of which 4277 mIU/L
(140 µg/L) was monomeric. One month later, her serum prolactin
(Bayer Immuno 1) was 17 660 mIU/L (600 µg/L), similar to the
concentration measured when she was first investigated.
The data presented on this patient suggest that the macroprolactin had some typical characteristics in that it is a prolactin-IgG antibody complex that is not bioactive in vivo. It is, however, unusual in being extensively glycosylated with, consequently, higher molecular mass and markedly different reactivity with commonly used immunoassay systems. A patient with extensively glycosylated high-molecular mass prolactin was reported by Hattori (5), but an IgG component was not demonstrated by protein-G affinity chromatography and there was no increased binding of 125I-labeled prolactin.
Hyperprolactinemia attributable to macroprolactinemia is commonly found in patients with moderate but not markedly increased serum prolactin concentrations [usually <3000 mIU/L (100 µg/L)]. Serum prolactin concentrations >6000 mIU/L (200 µg/L) are commonly taken as evidence of a prolactin-secreting pituitary adenoma (7). Others investigators (8) have reported the presence of big (as opposed to macro-) prolactin in two men with pituitary adenomas. Big, big prolactin, or macroprolactin, has been described as the major immunoreactive prolactin species in the serum of several subjects before and during pregnancy (5)(9)(10)(11).
In the case we report, monomeric prolactin increased, but macroprolactin remained the major immunoreactive prolactin component in serum during pregnancy. Given the medical history and subsequent progress of this present patient, it is unlikely that the vast majority of her circulating prolactin (up to 90% macroprolactin) was bioactive.
The data presented on this patient confirm that, depending on the assay system used, widely varying estimates of serum prolactin concentrations may be encountered in serum containing macroprolactin. It seems probable that the varying response of assay systems reflects the variable structure of macroprolactin and the availability of epitopes on the prolactin component to react with assay antibodies.
|
References |
|---|
|
TopIntroductionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  C. Hekim, H. Alfthan, J. T. Leinonen, and U.-H. Stenman Effect of Incubation Time on Recognition of Various Forms of Prolactin in Serum by the DELFIA Assay Clin. Chem., December 1, 2002; 48(12): 2253 - 2256. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) treatment with
4 mol/L urea. (B), gel filtration chromatography of
serum prolactin in the index case before (+) and after (*) treatment
with 4 mol/L urea.