INFORMATION TO USERS
This was produced from a copy of a document sent to us for microfilming. While the most advanced technological means to photograph and reproduce this document have been used, the quality is heavily dependent upon the quality of the material submitted.
The following explanation of techniques is provided to help you understand markings or notations which may appear on this reproduction.
1. The sign or "target” for pages apparently lacking from the document photographed is "Missing Page(s)”. If it was possible to obtain the missing page(s) or section, they are spliced into the film along with adjacent pages. This may have necessitated cutting through an image and duplicating adjacent pages to assure you of complete continuity.
2. When an image on the film is obliterated with a round black mark it is an indication that the film inspector noticed either blurred copy because of movement during exposure, or duplicate copy. Unless we meant to delete copyrighted materials that should not have been filmed, you will find a good image of the page in the adjacent frame. If copyrighted materials were deleted you will find a target note listing the pages in the adjacent frame.
3. When a map, drawing or chart, etc., is part of the material being photo graphed the photographer has followed a definite method in “sectioning” the material. It is customary to begin filming at the upper left hand corner of a large sheet and to continue from left to right in equal sections with small overlaps. If necessary, sectioning is continued again—beginning below the first row and continuing on until complete.
4. For any illustrations that cannot be reproduced satisfactorily by xerography, photographic prints can be purchased at additional cost and tipped into your xerographic copy. Requests can be made to our Dissertations Customer Services Department.
5. Some pages in any document may have indistinct print. In all cases we have filmed the best available copy.
University Microfilms International 300 N. ZEEB RD., ANN ARBOR, Ml 48106 8121836
N g , K w o kei Ja c o b
GAS CHROMATOGRAPHIC-MASS SPECTROMETRIC PROFILING OF ORGANIC ACIDS OF HUMAN AMNIOTIC FLUID
The Ohio Stale University PH.D. 1981
University Microfilms
Internationai300 N. Zeeb Road, Ann Arbor, M I 48106
Copyright 1981 by Ng, Kwokei Jacob All Rights Reserved GAS CHROMATOGRAPHIC— MASS SPECTROMETRIC PROFILING OF ORGANIC ACIDS OF
HUMAN AMNIOTIC FLUID
DISSERTATION
Presented in Partial Fulfillment of the Requirements for the Degree
Doctor of Philosophy in the Graduate
School of The Ohio State University
by
Kwokei Jacob Ng
•k ' k ' k & ' k
The Ohio State University
1981
Approved by
Reading Committee:
Brian D. Andresen , ; ■ / // / Joseph R. Bianchine /r"~ A 1'- A I Richard H. Fertel Advisor Philip B. Hollander Department of Pharmacology College of Medicine ACKNOWLEDGMENTS
Since the beginning of this project, the constant support, enthusiasm, advice and assistance in many ways from Dr. Bianchine and Dr. Andresen have enable me to accomplish the present work. It is also a great blessing and pleasure to have the opportunity to serve apprenticeship with Dr.
Andresen. He has spent many long late hours teaching me patiently the fine arts of gas chromatography and mass spectrometry, showing me how to apply the scientific method for clinical assays. The design of the helium preheater would not have been completed without his expertise.
I would like to thank members of the Reading Committee and Graduate
School Representative, Dr. Burkman for their time in reading my dissertation and suggestions. I would also like to thank Drs. Zuspan, 0'Shaunghessy, lams, Rayburn, Freeman and Stemple for their enthusiasm and supply of amniotic fluid samples and other biological specimens. Special thanks are due to Dr. Fertel who has been very helpful in many ways.
I would also like to thank Dr. Tejwani, Dr. Tjoe and Dr. Gerber for their interest in my research work. Mrs. Judy Lubbers has been very kind to share with me her capillary columns and experiences in handling them. The friendship of fellow graduate students is a cherisable memory.
Lastly, but not the least, the love and care of my family and my wife,
Shirley, have made the present effort the more worthwhile.
ii VITA
May 1, 1943...... Born - Canton, China
1963...... Northcote Teachers' Training College, Hong Kong
1963-1968...... School teacher, St. Louis School Hong Kong
1969-1973...... B.S., Department of Chemistry, California State University, Northridge
1973-1976...... M.S., Departments of Biochemistry & Pharmacognosy, Columbus, Ohio State University
1977...... Research Assistant I,II Department of Radiology, Columbus, Ohio
1978-1979...... Research Associate, Department of Radiology, Columbus, Ohio State University
1980-1981...... Ph. D., Department of Pharmacology, Columbus, Ohio State University
PUBLICATIONS
"Longevity, Stability and DNA Repair" by Hart, R.W., D'Ambrosio, S.M., Ng, K.J., and Modak, S.P. Mechanism of Ageing and Development, 9, 203 (1979).
"Helium Preheater for the Open-split Interface" by B.D. Andresen, K.J. Ng, J. Wu and J.R. Bianchine. Biomedical Mass Spectrometry (in print) FIELDS OF STUDY
Major Field: Pharmacology
Others: Biochemistry. Drs. Serif, Barber, Gross, Behrman Ives, Royer
Radioisotope Methodology. Drs. Malspeis and Feller
Organic Chemistry. Drs. Newman, Paquette, Swenton
Carbohydrate Chemistry. Dr. Horton
Natural Product Isolation Techniques. Dr. Doskotch
Medical Physiology. Drs. Grossie, Lipsky, Weiss,
Dujardin
Biochemical Methods. Drs. Ives, Behrman, Gross
Methods in Organic Chemistry. Dr. Ouellette
Gas Chromatography and Mass Spectrometry. Drs. Andresen, Hammar, Wong TABLE OF CONTENTS
Page ACKNOWLEDGMENTS...... ii
VITA...... iii
LIST OF TABLES...... vii
LIST OF FIGURES...... viii
I. INTRODUCTION...... 1
A. Introduction...... 1
B. Funtions of Amniotic fluid...... 2
C. The origin of amniotic fluid...... 2
D. Amniocentesis...... 5
E. Prenatal diagnosis by amniocentesis...... 8
F. Components of amniotic fluids...... 16
1. Cellular...... 16
2. Macromolecules...... 16
3. Organic acids...... 21
4. Organic bases...... 44
5. Hormones...... 46
6. Drugs...... 48
G. GC-MS Methodology...... 50
H. Objectives of present investigation...... 56
II. Materials and Methods
A. Biological samples...... 57
v Page
B. Extraction of amniotic f l u i d ...... 58
C. Derivatization of acidic fractions...... 62
D. Gas chromatography-mass spectrometry analysis of samples...... 65
III. Results
A. Stabilization of baseline with open-split interface heater...... 70
B. Cleanliness of instruments and purity of reagents...... 74
C. Capillary GC-MS profiles of acidic fractions of
1. amniotic fluid of early pregnancy...... 82 2. amniotic fluid of late pregnancy...... 86 3. amniotic fluid of pregnancy with open tubular neural defects...... 88 4. amniotic fluid of pregnancies with Rh sensitization.. 89 5. amniotic fluid of pregnancy from smoking mother 92 6. amniotic fluid of pregnancy with Delalutin treatment.. 94 7. amniotic fluid of pregnancy with hydrops...... 96 8. amniotic fluid of diabetic pregnancy...... 97 9. amniotic fluid from pregnancy of Down syndrome and gastrointestinal block...... 99 10. amniotic fluid from drug addicted mother...... 100
D. Capillary GC-MS profile of acidic fraction of serum...... 101
E. Capillary GC-MS profile of acidic fraction of ovarian cyst fluid...... 103
F. Comparison of organic acids identified in this study with those identified by GC-MS profiling of amniotic fluid in the literature...... 105
IV. Conclusions...... 109
V. Bibliography...... 115
VI. Appendices
Appendix A. Compilation of capillary GC-MS profiles of basic fractions of amniotic fluid...... 124
Appendix B. Compilation of mass spectra...... 139
vi LIST OF TABLES
Examples of genetic disorders and the corresponding diagnostic tests...... 9
Proteins, peptides and enzymes in amniotic fluid...... 20
Lists of organic acids in amniotic fluid identified by GC-MS profiling in literature...... 30
Steroidal hormones and metabolites in amniotic fluid.. 46
Non-steroidal hormones in amniotic fluid...... 47
Drugs and exogenous chemicals detected in amniotic fluid...... 48
List of capillary GC-MS analyzed samples and patient informations...... 80
List of identified compounds with the corresponding peak numbers...... 81
Comparison of organic acids identified in the literature with those of current investigation...... 107 LIST OF FIGURES
Figure Page
1 Extraction Flow Chart 1...... 59
2 Extraction Flow Chart 2...... 61
3 -Diazomethane micro-generator...... 64
4 Helium preheater and open-split oven (top view)...... 72
5 Helium preheater and open-split oven (end view)...... 73
6 Enlargement of open-split interface with alignment jacket...... 73
7 Comparison of TI plots with/without helium preheater.... 75
8 TI. plots of acidic, basic and neutral fractions of amniotic fluid (packed column)...... 76
9 System and solvent check...... 77
10 TI plot of trimethylsilylated CH2CI2 extract of acidic components of AF(E1): (15.5 weeks, tapped for maternal age)...... 82
11 TI plot of trimethylsilylated CH2CI2 extract of acidic components of AF(E1): 16 weeks, tapped for maternal age)...... 84
12 TI plot of trimethylsilylated CH2CI2 extract of acidic components of AF(E1): (38.5 weeks, no medication, repeat Cesarian section)...... 86
13 TI plot of trimethylsilylated CH2CI2 extract of acidic components of AF(E1): (15 week, open tubular neural defects)...... 88
14 TI plot of trimethylsilylated CH2CI2 extract of acidic components of AF(E1): (30 weeks, severe Rh-immunized, Phenytoin 100 mg bid)...... 89
viii Figure Page
15 TI plot of trimethylsilylated CH2CI2 extract of acidic components of AF(E1): (37 weeks, Rh sensitized, OD45Q = 0.01)...... 91
16 TI plot of trimethylsilylated CH2CI2 extract of acidic components of AF(E1): (16-18 weeks, nicotine 0.5 pack per day, Tylenol, vitamins) 92
17 TI plot of trimethylsilylated CH2CI2 extract of acidic components of AF(E1): (38 weeks, Delalutin once a wk, tapped for L/S)...... 94
18 TI plot of trimethylsilylated CH2CI2 extract of acidic components of AF(E1): (hydrops)...... 96
19 TI plot of trimethylsilylated CH2CI2 extract of acidic components of AF(E1): (35 weeks, diabetic).. 97
20 TI plot of trimethylsilylated CH2CI2 extract of acidic components of AF(E1): (Down syndrome and GI block)...... 99
21 TI plot of trimethylsilylated CH2CI2 extract of acidic components of AF(E1): (pentazocine, malnutrition, anemia)...... 100
22 TI plot of trimethylsilylated CH2CI2 extract of acid components(El) of serum from pregnancy with open tubular neural defects...... 102
23 TI plot of trimethylsilylated CH2CI2 extract of acidic components(El) of ovarian cyst fluid 104
24 TI plot of trimethylsilylated CH2CI2 extract of basic components of AF (15.5 weeks, tapped for maternal age)...... 125
25 TI plot of trimethylsilylated CH2CI2 extract of basic components of AF (16 weeks, tapped for maternal age)... 126
26 TI plot of trimethylsilylated CH2CI2 extract of basic components of AF (38.5 weeks, no medication, repeat Cesarian section)...... 127
27 TI plot of trimethylsilylated CH2CI2 extract of basic components of AF (open tubular neural defect)...... 128
ix Figure Page
28 TI plot of trimethylsilylated CH2CI2 extract of basic components of AF (30 weeks, severe Rh)...... 129
29 TI plot of trimethylsilylated CH2CI2 extract of basic components of AF (37 weeks, Rh sensitized, OD450 = 0.01)...... 130
30 TI plot of trimethylsilylated CH2CI2 extract of basic components of AF (16-18 weeks, nicotine 0.5 pack per day, Tylenol, vitamins)...... 131
31 TI plot of trimethylsilylated CH2CI2 extract of basic components of AF (38 weeks, Delalutin once/week, tapped for L/S)...... 132
32 TI plot of trimethylsilylated CH2CI2 extract of basic components of AF (Hydrops)...... 133
33 TI plot of trimethylsilylated CH2CI2 extract of basic components of AF (35 weeks, diabetic)...... 134
34 TI plot of trimethylsilylated CH2CI2 extract of basic components of AF (Down syndrome and GI block)...... 135
35 TI plot of trimethylsilylated CH2CI2 extract of basic components of AF (drug addict)...... 136
36 TI plot of trimethylsilylated CH2CI2 extract of basic components of serum (open tubular neural defect)...... 137
37 TI plot of trimethylsilylated CH2CI2 extract of basic components of ovarian cyst fluid...... 138
38 mass spectrum of lactic acid di-TMS...... 140
39 mass spectrum of 2-OH-butyric acid di-TMS...... 141
40 mass spectrum of 3-OH-butyric acid di-TMS...... 142
41 mass spectrum of 3-OH-isobutyric acid di-TMS...... 143
42 mass spectrum of 2-OH-2-methylbutyric acid di-TMS...... 144
43 mass spectrum of acetoacetic acid di-TMS...... 145
44 mass spectrum of benzoic acid TMS...... 146
45 mass spectrum of urea di-TMS...... 147 x mass spectrum of 2-keto-isocaproic acid di-TMS..... 148
mass spectrum of 2-0H-caproic acid di-TMS...... 149
mass spectrum of 2-keto-valeric acid di-TMS...... 150
49 mass spectrum of 2-keto-isovaleric acid di-TMS..... 151
50 mass spectrum of octanoic acid TMS...... 152
51 mass spectrum of phenylacetic acid TMS...... 153
52 mass spectrum of phosphoric acid tri-TMS...... 154
53 mass spectrum of 2-keto-3-methyl-valeric acid di-Tms 155
54 mass spectrum of 2-keto-caproic acid di-TMS...... 156
55 mass spectrum of nonanoic acid TMS...... 157
56 mass spectrum of decanoic acid TMS...... 158
57 mass spectrum of phenyl-d5~mandelic acid di-TMS.... 159
58 mass spectrum of salicylic acid di-TMS...... 160
59 mass spectrum of phenyllactic acid di-TMS...... 161
60 mass spectrum of 4-acetylaminophenol di-TMS...... 162
61 mass spectrum of p-OH-phenylacetic acid di-TMS..... 163
62 mass spectrum of lauric acid TMS...... 164
63 mass spectrum of homovanillic acid di-TMS...... 165
64 mass spectrum of hippuric acid di-TMS...... 166
65 mass spectrum of hippuric acid TMS...... 167
66 mass spectrum of myristic acid TMS...... g o ...... 168
67 mass spectrum of vanillylpropionic acid di-TMS..... 169
mass spectrum of indole-3-acetic acid di-TMS...... 170
mass spectrum of pentadecanoic acid TMS...... 171
mass spectrum of palmitoleic acid TMS...... 172
xi Figure Page
71 mass spectrum of palmitic acid TMS...... 173
72 mass spectrum of linoleic TMS...... 174
73 mass spectrum of oleic acid TMS...... 175
74 mass spectrum of stearic acid TMS...... 176
75 mass spectrum of squalene...... 177
76 mass spectrum of cholesterol TMS...... 178
77 mass spectrum of caffeine...... 179
78 mass spectrum of cotinine...... 180
xii I. INTRODUCTION
A. Introduction
Interest in human amniotic fluid dates back to the days of
Hippocrates. He first suggested that it is a product of fetal kidneys
(McCarthy and Sanders). In 1838, Rees described the presence of urea,
sodium, chloride and albumin in amniotic fluid. Parvin (1887) reported
the same biochemical constituents and noted also the presence of cells.
He speculated that the composition of amniotic fluid may vary as preg
nancy progresses. In 1919 Uyeno studied some physical properties and
chemical compositions of human amniotic fluid. During the last two
decades it was established that the composition of amniotic fluid
changes with gestational period, e.g. osmolality and sodium concentra
tion diminishes as the fetus develops while the concentrations of urea
and creatinine increase (Pitkin and Zwirek, 1967; Gillibrand, 1969;
Lind et al, 1972). With the increasing availability of amniotic fluid
during recent years, because of the increasing utilization of amniocen
tesis, there is growing interest in its origins, functions, composition,
and its dynamic changes throughout the gestational period. The under
standing of the nature of amniotic fluid in relation to the growth and
development of the conceptus has provided invaluble informations in
prenatal diagnosis and perinatal care. Various aspects of amniotic
fluid will first be reviewed.
1 B .____Functions of Amniotic Fluid 2
Amniotic fluid is enclosed within the amniotic sac in which the fetus is suspended. This fluid provides a protective cushion against
jolts by absorbing and distributing impacts the mother may receive.
It allows the embryo to grow, develop symmetrically, and prevents the amnion from adhering to the sticky embryo. The fluid helps to control the embryo's body temperature at a relatively constant point.
It enables the fetus to move freely, thus aiding musculoskeletal development by providing a semi-weightless environment.
C. The Origin of Amniotic Fluid
The amnion is the innermost of the two human fetal membranes.
Hertig et al (1956) described how on the eighth day after ovulation the amniotic cavity emerges between the trophoblast and the inner cell mass which develops to form the bilaminar embryonic disc composed of the
embryonic ectoderm and embryonic endoderm. As the amniotic cavity
enlarges it acquires a thin epithelial roof, the amnion, probably from
the cytotrophoblastic cells. The embryonic ectoderm forms the floor of
the amniotic cavity and is continuous peripherally with the amnion.
The cavity is filled with fluid and initially is called the embryonic vesicle which is observable microscopically in all ova which have
embedded in the decidua (Rhodes 1969). As the embryonic disc develops,
the rapidly enlarging amniotic sac surrounds the fetus forming a
sheath for the umbilical cord. Eventually the chorionic cavity is
completely obliterated by the growth of the amniotic sac and the amnion
is then in direct contact with the chorionic membrane (Moore, 1973;
McCarthy and Saunders, 1978). 3 Undoubtedly, amniotic fluid must initially originate from the mother. However, the exact physiological processes involved in its production are still not clear. This lack of understanding is due to the multiple sites involved, and their growth variation as pregnancy progresses. Immunological studies have shown that amniotic fluid con tain proteins derived both from fetal and maternal sources (Sutcliffe,
1975). The amniotic fluid could arise from a variety of sources, including 1) an ultrafiltrate from the embryonic epithelium (or fetal skin), 2) an ultrafiltrate from maternal circulation, 3) a filtrate from the placenta, umbilical cord and amniochorionic membranes, 4) urine from kidneys, or 5) secretions of the tracheobronchial tree or gastro intestinal tract.
During the first half of pregnancy amniotic fluid appears to be derived almost entirely as a filtrate of fetal serum (Lind, 1978).
The concentrations of the more diffusible solutes in amniotic fluid
(e.g. sodium, chloride and urea), are more closely related to those of
fetal than maternal serum. Lind and Hytten (1970) suggested that amniotic fluid may be regarded as part of the fetal extracellular fluid
space. There is a close correlation between fetal weight and amniotic
fluid volume. During the second half of pregnancy, fetal skin becomes
keratinized and relatively impermeable, the composition of amniotic
fluid deviates progressively from those of fetal or maternal serum
toward fetal urine. There are marked differences during pregnancy in
the permeability of fetal skin in vitro (Parmley and Seeds, 1970; Lind
et al., 1972) and in vivo (Abramovich and Page,1972). There is a
specialized area of in the part of the amnion overlying the placenta 4 which, because of its extensive vascular connections, plays a great part in fluid exchange. Vosburgh et al (1948) and Hutchinson et al
(1955) demonstrated by isotope studies that there is fluid exchange between the fluid compartments of the fetus and mother. Campbell et al (1973) and Dawes (1974) measured fetal kidney and lung functions respectively, by ultrasonic methods.
The cause of polyhydramnios, the presence of abnormally large volume of amniotic fluid, is not clear and apparently contradicting phenomena have been observed. For example, polyhydramnios has been associated with fetal bilateral renal agenesis (Potter, 1965). However, normal volumes of amniotic fluid (Taussig, 1927; Sylvester and Hughes,
1954) were also reported in cases of renal agenesis (Jeffcoate and
Scott, 1959; Bain and Scott, 1960). Similarly, instances of anen- cephaly and agenesis of fetal gastrointestinal tract have been associ ated with polyhydramnios but they have also occurred with normal volume of amniotic fluid.
The classical way of measuring amniotic fluid is by collecting the fluid at the time of rupture of the membranes or by removal of the intact sac at hysterotomy. The volume of amniotic fluid is generally obtained by a dilution method (Dieckmann and Davis, 1933), using sodium aminohippurate. Abramovich (1970) found that the gestational age (10-20 weeks), as measured from the last menstrual period, correlated
significantly with the logarithm of the liquor volume, with a correla
tion coefficient of 0.78. The regression equation is:
y = 0.117 x -I- 0.36 (1) 5 where y = log of liquor volume and x = gestational age in weeks.
There were also significant direct correlation between the volume of amniotic fluid and fetal crownrump length, fetal weight and placental weight (Abramovich 1968). Charles et al (1965) investigated 27 mid trimester pregnancies and in over 80% of these cases, the volume ranged from 482-785 ml. Queenan et al (1972) found that the volume averages were 669 ml at 25-26 weeks of gestation, 948 ml at 33-34 weeks and 836 ml at term. However, there was a great variation in volume between individual patients.
The origins of amniotic fluids and its exact route of circulation at various stages of pregnancy are not yet clear. It is known that both the mother and fetus contribute to the fluid and solutes of the amniotic fluid. The fetal skin, amniotic cells, fetal cord, placenta, fetal swallowing and urination play significant roles during the various stages of pregnancy. The mechanisms of controlling the volume and composition need to be more fully investigated.
D. Amniocentesis
Amniocentesis concerns the insertion of a needle into the amniotic cavity. The procedure is usually performed transabdominally, but a vaginal approach has also been used. Amniotic fluid may be removed or, conversely, medications or dyes can be introduced. In 1930 Menees and associates first advocated amniocentesis as a diagnostic aid in practical obstetrics. Amniocentesis was used for the injection of hypertonic saline to induce therapeutic abortion. During the past 25 years, amniocentesis has been most indicated for the management of pregnancies complicated by Rhesus isoimmunization.
In the vast majority of cases amniocentesis is indicated for 6 diagnostic purposes. These could be due to the advanced age of the mother, familial history of birth defect, abnormalities in previous child or children, or exposure to physical or chemical mutagens or teratogens. According to the studies done by the Medical Research
Council of London (1978), among 3131 amniocenteses performed, 52.1% were indicated to detect chromosomal abnormalities, 40.9% to measure alpha-fetoprotein as a screen for neural tube defects, while the re maining 7% involved miscellaneous reasons. Amniocenteses performed during the third trimester usually assess fetal pulmonary maturity or aid in the management of pregnancies complicated by diabeted mellites or Rhesus isoimmunization.
Three separate groups conducted programs to evaluate the safety of the procedure. They were the U.S. National Institue of Child Health and
Human Development (N1CHHD) Study Group, 1976; Simpson et al in Canada,
1976 and the Working Party on Amniocentesis in the United Kingdom. It is difficult to compare directly the numerical findings between these studies because of variations in the selection of the subjects who underwent amniocenteses and variations of the control ''roup (when present). However, all studies consistently demonstrated an increase of fetal loss of no more than 1% as a result of amniocentesis. Fetal loss is defined as the total number of pregnancies which do not result in a live baby, thus including spontaneous abortion, still birth, intrauterine death, and perinatal death. The fetal losses of the
NICHHD, Canadian and MRC studies are 3.46%, 4.46%, 2.6% in the subject group and 3.21%, 3.35%, 1.16% in the control group, respectively.
Compared with the controls, there are in the subject group an as yet unexplained increase in the incidence of respiratory distress in the 7 newborns and also an increase in the incidence of major orthopedic defor mities (e.g. club foot). The latter this might possibly be related to removal of amniotic fluid at a critical period jeopardizing the sym metrical development of the feet. It might be worthwhile to determine whether or not those mothers with amniocentesis who delivered infants with severe orthopedic abnormalities corresponded to those pregnancies in which there was a lower than normal volume of amniotic fluid.
Other complications which have been observed include antepartum haemorrhage from both placental praevia and abruptio placentae, blood stained fluid, fetal-maternal transfusion, vaginal leakage of amniotic fluid, premature labour, and chorioamnionitis. Recent reports show that the hazards vary with indication. If amniocentesis is performed because of a higher level of alpha-fetoprotein in maternal serum, it is associated with higher rate of fetal loss than performed for any other reason (Bennett, 1978; Weiss et al 1979). Further studies are required to define the two complications of amniocentesis.
The safety of amniocentesis can be improved by taking appropriate precautions. Ultrasonic scans can provide visualization of the position of the placenta, the position of the anatonmy of the fetus, and the location of the pool of amniotic fluid. The procedure should be avoided before the 16th week of gestation. Amniocentesis should be carried out immediately after ultrasonic examination, preferably without further moving the patient. It would be better if real-time ultrasonic moni toring is available. Careful selection of site, proper angle and depth of insertion of needle, routine use of sterile procedure, sharp needle with a stylet to prevent coring out of the maternal tissue, and needle of a size no large than gauge 20, all have helped to diminish 8 the hazards of amniocentesis. Further, Rhesus negative woman without
Anti-D antibodies should receive a dose of Anti-D immunoglobulin.
The procedure should be done at 'genetic centers' where adequate facilities and experienced personnel are available. While maternal complications due to amniocentesis are rare, there is no doubt that natural hazards with amniocentesis do occassionally occur. Judgement has to be made between the hazard of the procedure to fetus and mother and the value of defining the existence of fetal malformations to the decision of possibly terminating the pregnancy.
E. Prenatal Diagnosis By Amniocentesis
The list of genetic disorders or adverse fetal conditions detectable
by amniocentesis will undoubtedly lengthen with time as more biomedical
and technological discoveries are mad-a. It is not intended here to
review all prenatal diagnostic applications of amniocentesis. However,
common applications will be discussed.
Chromosomal Disorders
One of the most common indications of amniocentesis concerns the
detection of chromosomal abnormalities associated with advanced mater
nal age. Viable cells derived from amniotic fluid can be cultured and
allowed to multiply to give enough cells for chromosomal analysis.
Various chromosomal aberrations can be observed when cell growth is
arrested at metaphase, and the cells duly processed and stained. For
example, genetic disorders due to trisomy, different type of transloca
tions, deletions, or breakage of chromosomes can be identified. Down's
syndrome is a genetic disorder due to trisomy 21. The affected indivi
dual has characteristic features of head, nose, hands and feet, and
various degree of mental retardation. The incidence varies with maternal age, ranging 1:2000 for those under 29, to 1:50 after 45 years of age (Carr, 1969, Valenti et al, 1969).
Other genetic disorders that have been prenatally diagnosed have been recently reviewed (Patrick, in print). A few examples are listed below:
Table 1. Examples of Genetic Disorders and Corresponding Diagnostic Tests.
Disease Diagnostic Test
Tay-Sachs hexosaminidase A Sandhoff hexosaminidase A and B Gml~gangliosidosis beta-galactosidase Hurler alpha-iduronidase methylmalonic acidemia ®12-resPonsi-ve deoxyadenosyl~Bi2 synthesis B^-non-responsive methylmalonyl-CoA mutase glycogen storage type II alpha-glucosidase pyruvate decarboxylase deficiency pyruvate decarboxylase congential nephrosis alpha-fetoprotein
Neural Tube Defects
Neural tube defects (NTD) occur in about one per 1000 births. The frequency varies with locality, ethnic groups and socio-economic classes.
They are due to defective development of the neural tube during the third and fourth weeks of gestation, resulting in (a) the absence of the cranial vault with cerebral hemispheres completely missing or reduced to small masses as in the case of anencephaly or, (b) defective closure of the bony encasement of the spinal cord as in the case of spina bifida. Other variations of neural tube defects include inence- phaly, spina bifida with meningocele, myelomeningocele, myeloschisis
(open tube) and hydrocephaly. Some malformations are due to genetic factors while some result from environmental factors such as infectious 10 agents, drugs, ionizing radiations, and metabolic diseases. Some forms of neural tube defects are incompatible with life and those that survive often suffer severe handicaps such as musculoskeletal paralysis.
The measurement of the concentration of alpha-fetoprotein (AFP) in amniotic fluid was first used as a means of prenatal diagnosis of both anencephaly and open spina bifida in 1972 (Brock and Sutcliffe).
Although the concentrations of AFP in amniotic fluid obtained from prenancies afflicted with anencephaly or other open neural tube defects are higher than those from normal pregnancies, there is an overlap between the two groups. Furthermore, there are other fetal conditions which have been reported to associate with higher than normal concentra tion of AFP in amniotic fluid. These conditions include intra-uterine death, congenital nephrosis, exomphalos, fetal teratoma, polycystic kidneys, congenital skin defects, Turner's sysndrome, duodenal atresia, oesophageal atresia, Fallot's tetralogy and annular pancreas (Brock,
1979). Because of the low occurrence of NTD in the general population relative to the risks involved in amniocentesis, general screening of
NTD by measuring AFP in amniotic fluid is not feasible. According to the Report of the U.K. Collaborative Study on Alphafetoprotein in
Relation to Neural Tube Defects (1977), 90% of cases of both of anen cephaly and spina bifida would be detected at the 95th percentile of the normal range of AFP in maternal serum at 16-18 weeks of gestation.
Hence, measurement of maternal serum level of AFP can be used as a screening procedure and the measurement of AFP level in amniotic fluid as a diagnostic procedure. An inrease on maternal serum level of AFP was also associated with multiple pregnancies, intrauterine death and threatened abortion (Seppala and Ruoslaliti 1972). There has been a continous search for a more definitive marker of
NTD. Smith et al (1979), utilizing the heterogeneity of AFP suggested
that measuring the percentage of concanavalin-A-non-reactive AFP in
total amniotic AFP would be a valuable test for the detection of NTD.
Brock et al (1980) reported a method of distinguishing NTD and congeni
tal nephrosis of the Finnish type. While both defects have higher
than normal levels of AFP in amniotic fluid, the amniotic fluid from
NTD pregnancies contained an extra band of cholinesterase on polyacryl amide gel electrohoresis.
Ultrasonography is also useful in the prenatal diagnosis of NTD.
In summary, the search for a definitive marker and method for prenatal
diagnosis of NTD goes on.
Fetal Lung Maturity
The majority of indications for amniocentesis near term are for
the assessment of fetal pulmonary maturity, especially in the manage ment of pregnancies complicated by Rh isoimmunization or diabetes mellitus. Premature infants often suffer from respiratory distress
syndrome (RDS) with subsequent hyaline membrane disease. RDS was
associated with a mortality ratio of approximately 30%
along with significant risks of long-term neurological or pulmonary
disorders. The cause of the disorder is due to immaturity of the
lungs. Mature lungs are protected from collapsing on expiration by
surface active phospholipids synthesized in the lungs. Any pure water
layer has a tendency to contract into a minimum surface area because
of the cohesion between water molecules. The surface tension of water is about 72 dynes/cm and must be lowered below 15 dynes/cm to
sustain life (Clements et al, 1958). The lowering of surface tension 12 of water in the alveoli is achieved by the biological surfactants
which are found in the lamellar bodies in the type II alveolar cells.
As gestation progresses, increasing number of surfactant containing
lamellar bodies are extruded from the cytoplasm of type II alveolar
cells into alveolar fluid which is released into amniotic fluid (Strang,
1977).
The surfactants are made up of about 10% protein and 90% lipid.
About 90% of the lipid is phospholipids, which include lecithin
(phosphatidyl choline), acidic phospholipids, phosphatidyl inositol,
phosphatidyl glycerol, and sphingomyelin. Kulovich et al (1979) found
that lung maturity is reflected in the measurement of lecithin, phos
phatidyl inositol, phosphatidyl glycerol and sphingomyelin. Lung
maturity is also a reflection of the surfactant maturation process
during which the unsaturated fatty acid moiety on the lecithin molecule
is replaced by saturated fatty acid, usually palmitic acid. Saturated
lecithin has a greater surfactant activity. With unsaturated fatty
acids moiety on lecithin surface tension goes no lower than 23 dynes/cm.
When saturated acid is substituted, surface tension can be reduced to
0-4 dynes/cm.
Phosphatidyl glycerol, which begins to appear at about 36 weeks of
gestation, has some importance as a surfactant in fetal and neonatal
lungs. Hallman et al (1975) reported the abscence of phosphatidyl
glycerol during the active phase of RDS, but reappeared during recovery.
Many methods have been devised for the assessment of fetal pulmonary
maturity by measuring either the surfactants in amniotic fluid or the
physical properties of amniotic fluid due to their presence. Gluck et
al (1971) used the lecithin/sphingomyelin (L/S) ratio for the 13 diagnosis of respiratory distress syndrome. They determined that a
L/S ratio of 2.0 or greater denoted lung maturity in the newborns.
They also showed that when the fetus respiratory function becomes
'mature', there is not only a concomitant surge of lecithin in the amniotic fluid, but also a change in the metabolism of lecithin in the fetal lung, producing dipalmitoyl lecithin instead of lecithin with unsaturated fatty acids esterified to the C-2 position of the glycerol moiety of the molecule. The L/S ratio in amniotic fluid has been widely adopted in monitoring fetal lung maturity. Their method consists of centrifugation of amniotic fluid to remove cells and debris, extrac tion with chloroform-methanol, precipitattion with ice cold acetone, thin layer chromatographing of the acetone precipitate, spraying with sulfuric acid followed by charring and densitometric measurements.
Different variations of L/S ratio measurement have been recently reviewed (Wagstaff, 1978).
The major inadequacy of the L/S ratio of 2.0 as a criteria of fetal lung maturity is that it gives false negative values in the range between 1.5-2.0. Further, it also gives false positive data when pregnancies are complicated by diabetes raellitus.
Duck-Chong (1979) devised a micro-method for isolating a lung derived lamellar body fraction from amniotic fluid by means of isopynic density ultracentrifugation of whole amniotic fluid. The phosphorus content of the phospholipids from the lamellar body-rich fractions of
2 amniotic fluid samples were determined. Lecithins accounted for
72.3 and 76.1%; phosphatidylethanolamine and/or phosphatidylglycerol
11.9 and 13.0%; lysolecithin 2.4 and 3.8%; sphingomyelin 1.1 and 2.3%; 14 and phosphatidylserine and/or phosphatidylinositol 7.3 and 9.7%.
In a study involving 479 amniotic fluid samples from 334 pregnancies to compare the accuracies of assessing fetal lung maturity with the critical values L/S ratio of 2.0 and lamellar body phospholipid (LB-PL) concentration of of 3.5 mg/dl, Duck-Chong et al (1980) showed that
LB-PL values gave significantly less false negative results. While
44% of the samples having L/S ratio of less than 2.0 had LB-PL values of greater than 3.5 mg/dl, the latter seems to be a better evaluation of fetal lung maturity. They concluded the LB-PL method had the poten tial as a clinical procedure for determing fetal lung maturity.
However, more data need to be collected to establish this procedure.
Because the L/S ratio is unreliable for predicting RDS when the amniotic fluid sample is contaminated with blood, meconium, or if obtained during complicated pregnancies, Torday et all (1979) examined
322 amniotic fluid samples, 75% of which were contaminated as noted above or obtained during complicated pregnancies. They used both the
L/S method and a rapid technique involving osmium tetroxide developed by Mason et al (1976) for the isolation of saturated phosphatidyl choline (SPC), the major component surfactant. Of these 322 samples,
there were 45 cases of RDS; 25 of which were correctly predicated with
L/S ratios less than 2.0 (55.5%). However, using a saturated phospha tidyl choline (SPC) concentration of 500 ug/dl or less as criteria,
35 out of 42 cases of RDS (82%) were predicted. When the L/S ratio was greater than 2.0, there were 13 of 277 (4.7%) false lung maturity, whereas SPC gave 3 out of 280 (1.1%) false lung maturity prediction.
They concluded that the determination of SPC is both more specific and more sensitive as a predictor of RDS than other techniques currently 15 in use; especially for amniotic fluid samples contaminated with blood or meconium or obtained during complicate pregnancies which included premature rupture of membranes, erythroblastosis fetalis, abnormal placentation, diabetes mellitus (including 17 cases of polyhydramnios and oligohydramnios), preeclampsia or eclampsia, intrauterine growth retardation and fetal distress.
In 1958 Pattle showed that in air saturated water bubbles 50 um in diameter obtained from the lung may be stable for hours while ordinary bubbles of this size contract and disappear in a few minutes under the influence of surface tension. Based on this finding, Clements et al (1972) introduced a simple physical method for bedside prediction of fetal lung maturity. It is known as "foam stability test", "shake test", "bubble stability test (BST)" or "rapid surfactant test (RST)".
The procedures involved using a set of glass tubes labeled l-5 and of internal diameter of 8-14 mm and free of soap, serum or biological fluids. One ml of uncentrifuged amniotic fluid, free of blood or meconium contamination, was diluted with 0.9% saline solution at 1:1,
1:13, 1:2, 1:4, and 1:5 dilutions respectively. Then 1 ml of 95% ethanol was added to each of the tubes. The tubes were handshaken rapidly for 15 sec. After standing for 15 min the tubes were visually examined. When no complete circles of bubbles were seen, the result was considered negative (no lung maturity). A stable circle of bubbles in tubes 1 & 2 was considered as an intermdiate result. If a stable foam persisted in tubes 3, 4, or 5, then it was a positive result, denoting fetal lung maturity. Although unique, this foam stability test has the problem of giving false negative results.
Other methods for assessments of fetal lung maturity include the 16 measurement of palmitic acid concentration by gas chromatography (War ren et al 1974, Russsell et al 1974, Hood et al 1977, Bichler et al
1977), the palmitic acid/stearic acid ratio ( Zuspan et al 1975, O'Neil et al 1978, Lavoinne et al, 1980), enzymatic assay, radioimmunoassay and measurement of optical density of amniotic fluid at 450 and 650 nm.
F. Components of Amniotic Fluid
1. Cellular
Cells and cell debris float freely in amniotic fluid. Viable cells are identified by light microscope. These cells maintain their integral dynamic functions. They are able to exclude trypan blue stain, thus remain clear and refractile under the light microscope. Nonviable cells are stained blue. At 16 weeks of gestation 30-80% of all cells are alive. The total number of cells increases with gestation but the proportion of viable cells does not increase. At about 24 weeks of gestation, usually only 10-15% of the cells are alive (Gosden and Brock,
1977, 1978).
Cells in amniotic fluid vary in diameter from about 6 micron to 50 microns. At 16-24 weeks of gestation there are two major cell types, the large squamous cells of about 40-50 microns and the smaller round cells of about 15-25 microns. The larger ones are usually polygonal or ovoid, with non-vacuolated acidophilic or basophilic cytoplasm with vesicular or pyknotic nuclei. The smaller cells are round, ovoid or cuboidal with dense, vacuolated basophilic cytoplasm and a well defined nucleus in which a nucleolus is often prominent. These cells are believed to originate from the amnion (Van Leeuwen et al, 1965). 17 Study of the surface topography of individual amniotic fluid cells with a scanning electron microscope, defines more precisely the shape and size of cells along with the presence of prominent microvilli,
surface microtubules or surface convolutions. This information assists greatly in the determination of tissue origin, e.g. skin, amnion,
buccal mucosa or bladder lining (Gosden et al, in print).
Uncultured amniotic fluid cells can be used for detection of fetal
abnormalities. However, the amount of cells are small and generally
not sufficient for diagnostic purposes. In addition viable and non
viable cells may contain components of different biological properties,
not indicative of the fetus status.
Viable amniotic fluid cells placed into a culture medium with the
appropiate incubating environment (e.g. proper nutrients, pH, tem
perature, and atmosphere) will adhere to the culture plate and undergo mitosis. Cells from amniotic fluids of normal pregnancies usually
take three to four days to become attached. However, in some genetic
disorders amniotic fluids contain a large proportion of viable cells'
that can become attached to the culture plate within 12 to 24 hours or
so. These rapidly attaching cells (RAC) have morphologies distinct
from those of the cells from normal amniotic fluids. Gosden et al (in
print), described four abnormal cell types found in pregnancies of
neural tube defect. All are rapidly attaching cells and have the same
morphology as those from fetal brain or fetal spinal cord. In case of
spina bifida, RAC account for 950% of all viable cells, whereas in
anencephaly and spina bifida, RAC account for almost 100% of the viable
cells. There is another type of RAC which has the morphology of peri
toneal macrophage. They are found in exomphalos and account for 50% 18 of the viable cells. Placental cells found in amniotic fluid as a result of penetrating the anterior placenta during amniocentesis are also rapidly attaching and have their own distinct morphology. Gosden et al also noted that some amniotic fluid cells (which they called fetal distress cells (FDC)) are rapidly attaching and have characteris tic morphology. These FDC possibly are derived from fetal lung, trachea, bronchi, kidney, urogenital tract or upper layer of placenta. They are associated with impending abortion, pre-eclampsia, preterm birth and low birthweight. Hence, a combination of cellular adherence values, cell morphology and alpha-fetoprotein values can be used to identify some fetal abnormality.
By culturing amniotic fluid cells for two or more weeks, a larger number of cells can be obtained for chromosomal karyotyping, whereby the sex of the fetus as well as some chromosomal abnormalities can be identified. This provides invaluable informations to clinicians, genetic counsellors and parents.
2. Macromolecules
Macromolecules in amniotic fluid include nucleic acids, proteins, polysaccharides, lipids, lipoproteins and mucopolysaccharides. Such components, with the exception of some lipids, are currently not amen able to gas charomatographic-mass spectrometric-computer method of analysis because of their large molecular weights and non-volatility.
Field-desorption mass spectrometric method can currently handle mole cules of no greater than 2000 daltons. There is very little informa tion on nucleic acids in amniotic fluid in the biomedical literature, probably because there is very little free nucleic acid in amniotic fluid. However if they are present, current technology is not yet 19 capable of characterizing the nucleic acids in amniotic fluid in a
routine manner to yield useful information. Current examination of the genetic composition of the fetus is generally done on the chromosomal
level by culturing the cells in amniotic fluid followed by karyotyping.
No method is yet available to detect specific genetic defects by direct
examination of the sequence of DNA in the chromosomes.
There is a voluminous literature dealing with various proteins and
enzymes in amniotic fluid. These topics have been recently reviewed
by Queenan (1978) and Burton et al (1978). The most studied protein
is the alpha-fetoprotein in connection with neural tube defects. It
is beyond the scope of the present writing to review the proteins and
enzymes in amniotic fluid here. However, a listing of these macro molecules in amniotic fluid is presented in Table 2. Table 2. Proteins , Peptides, and Enzymes in Amniotic Fluid acetylcholinesterase glutaryl Co-A dehyrogenase acetylgalactosamine-6-sulfate- glycerophosphate phosphatase sulfatase glycoproteins acid esterase glycosidase acid glycoprotein glycosyltransferase acid hydrolase adenosine deaminase hydrolase agglutinin Ig agglutininogens A,B iduronate sulfatase albumins immunoglobulins (alpha-1, alpha- alkaline phosphatase beta-1, beta-2, A, G) alpha-fetoprotein isoamylase alpha-globulin alpha-2-macroglobulin kallikrein amylase kinin aminopeptidase kininase antigen kininogens antitrypsin arginosuccinase lactate dehydrogenase arylaminopeptidase lactalbumin arylsulfatase A leukemia differentiating factor aspartate aminotransferase leucyl aminopeptidase lipoprotein lysozymes bacterial aggregating factor bacteriolysins mannosidase beta-acetylhexoaminidase methemalbumin beta—endorphin methemoglobin beta-glucuronidase microglobulin bradykininogen monoamine oxidase muramidase carcinoembryonic antigen myoglobulin ceruloplasmin cholinephosphotransferase pepsin collagen pepsinogen cobalophilin peptidase creatine kinase phosphatases creatine phosphokinase phosphatidate phospho-hydrolase cystathionine synthetase phospholipase A2 prealbumin diamine oxidase procollagen pyruvate kinase eutocine fucosidase sphingomyelinase sialoglycoprotein galactosyltransferase gamma-glutamyl-transpeptidase transferrin gamma-globulins glutamic-oxalacetic transaminase 21 3. Organic Acids
Organic acids are important components in all living organisms, and their functions are numerous. One of the most important function
is to serve as building blocks of various cellular structural and
functional components. They also play an important role as intermediate
substrates in catabolic and anabolic pathways of fats, proteins and
carbohydrates, in the synthesis and degradation of cellular constituents
in the production of energy for the organism, and in the synthesis and degradation of molecules which regulate cellular activities.
Various organic acids in amniotic fluid have been studied in the
search for a better understanding of the conditions of growth and deve
lopment of the fetus, its clinical status and its maturity.
Fatty Acids
Fatty acids in amniotic fluid are of great interest to obste
tricians and perinatologists because they are one the chief building
blocks of phospholipids whose quantity and quality serve as an indicator
of fetal lung maturity. In the search for reliable parameters that
predict the maturation status of fetal lung, investigators have examined
free fatty acids in amniotic fluid and the fatty acid composition of
various lipid classes in human amniotic fluid, namely, phospholipids,
triglycerides, diglycerides, monoglycerides, cholesterol esters and
ether lipids. In 1973 Biezenski studied the esterified and unesterified
fatty acids in 8 samples of normal amniotic fluid at term. The fatty
acids quantitated by gas chromatography included lauric, myristic, myris
toleic, palmitic, palmitoleic, stearic, oleic, linoleic, linolenic,
arachidic and arachidonic acids. Palmitic acid was found to be the
most abundant fatty acid in each of the lipid classes (phospholipids, 22 cholesterol esters, mono-, di- and triglycerides, and unesterified free
fatty acids). Palmitic acid accounted for 53% and 37% of all fatty acids in phospholipids and in total free fatty acids, respectively.
Oleic acid is the second most abundant fatty acid in various lipid
classes with the exception of cholesterol esters. Biezenski reported
that while arachidonic acid was found both in phospholipids and as
free acid fractions, it was not esterified with cholesterol or with
glycerol. Arachidonic acid was found to be esterified with only
polar lipids and not with neutral lipids.
Das et al (1975) reported a gestational variation of fatty acid
composition of human amniotic fluid lipids. A total of 13 samples were
divided into 4 gestational groups: 18-22 weeks (3 cases); 27-33 weeks
(4 cases); 34-40 weeks (3 cases) and labor at term (3 cases). For each
of the gestational groups, the fatty acid composition of the following
lipid classes were quantitated: total lipids, cholesterol esters, trigly
cerides, free fatty acids, sphingomyelin and lecithin. The fatty acid
composition of each lipid class included all of the following 13 fatty
acids: pentadecenoic, palmitic, palmitoleic, heptadecanoic, stearic,
oleic, linoleic, linolenic, arachidic, arachidonic, behenic, docosapen-
taenoic and lignoceric. Total lipids increased with gestation,
especially at 34-40 week and at term. Palmitic, stearic and oleic were the major fatty acids in the total lipids at all stages of gesta
tion. There were no significant variation in fatty acid composition
in total lipids between the various gestational stages with the excep
tion of palmitic and heptadecanoic acids. However, at term the fatty
acid composition for cholesterol esters revealed a pattern of the
relative concentrations of the individual fatty acids that was distinct 23 from that at any other stage of pregnancy. In fact, cholesterol esters
of arachidonic, behenic and docosapentenoic acids increased with the advancement of gestation. They suggested that the level arachidonic acid in cholesterol esters might be useful in assessing fetal lung maturity. The percentages of arachidonic acid in the four stages of
gestation (18-22, 27-33, 34-40 wks, & at term) were 1.7 + 0.08, 2.6
+ 0.1, 5.9 + 0.1 and 6.9 + 0.1 respectively.
The major components of free fatty acids are palmitic, stearic and
oleic acid. Free stearic acid appeared to decrease at term while oleic acid increased at 34-40 weeks and at term. Total free fatty acids were
shown to increase with gestation (3.05 + 0.06%, 5.19 + 0.08%, 5.88 +
0.09%, 5.83 + 0.11%) in disagreement with results reported by Hagenfeldt
and Hagenfeldt (1976). As for sphingomyelin, the arachidonic acid content was very low at early pregnancy and unusually high at term (3.3 +
0.07%, 1.5 + 0.0%, 1.7 + 0.06%, 30.4 + 4„2%). As for lecithin,
the palmitic acid content increased with gestation, and the increase
became greatly accelerated at 34 weeks of pregnancy. The value of free
fatty acid profiles would be greatly enhanced if data from a larger
sample size could be obtained.
Hagenfeldt and Hagenfeldt (1976) measured by gas chromatographic
method individual free fatty acids in 77 samples of amniotic fluid and
57 samples of maternal plasma. Gestational age ranged from 16 to 40
weeks. They found that free fatty acids decreased from 20.4 + 6.0
umole/1 in the second trimester to 17.0 + 6.5 umole/1 in 34-40 weeks
(p<0.05) while data from Das et al (1975) showed an increasing trend.
The concentration of free stearic acid decreased while oleic acid 24 increased slightly with gestation, in agreement with the result of Das et al. Serial amniotic fluid samples showed that free fatty acids decreased with gestation. The concentration of free fatty aicds in maternal plasma was much higher than that in amniotic fluid of corres ponding gestation period in the range of 16-40 weeks. The fatty acid profiles also differed significantly between maternal plasma and amnio tic fluid, the latter containing more saturated fatty acids, particularly palmitic acid. Also, fatty acid composition of amniotic fluid resembled closely that of urine.
Schirar et al (1975) measured the fatty acids hydrolyzed from total lipid extract of 201 samples of amniotic fluid. Palmitic acid increased steadily up to 33-34 weeks of gestation. Then it rose abruptly to 53% at weeks 37 and 38. Palmitoleic acid, though in lower concentration, showed a similar trend. Stearic, oleic, linoleic and arachidonic acids de creased with gestation. Before week 36, 95% of the palmitic acid/stearic acid ratio (P/S) values were less than 5, and only 5% (6/91) were over 5; beyond week 36, 94% of the P/S ratio values were over 5, and only 6%
(5/110) were under 5. They also showed that in the group of premature infants with respiratory distress syndrome the percentage of palmitic acid in total lipid is much lower (34.18%) and significantly different
(p<0.002) than that found in infants delivered at term (52%) or healthy premature neonates (46.5%, p<0.05).
Balint et al (1978) compared the fatty aid compositions of lecithins from gastric aspirates, tracheal washings and amniotic fluid from infants with respiratory distress syndrome (RDS) with those from normal healthy infants. They found that the fatty acid composition of lecithins from gastric aspirates, tracheal washing or amniotic fluid of infants with 25 RDS was similar. However, when compared to either full term or pre mature control subjects without RDS, palmitic acid was decreased, and stearic acid content increased. When the molar percentage of palmitic acid in lecithin from gastric larvage was above 60% none of the infants developed RDS, whereas 10 of 11 with palmitic acid concentration of
50% or less did developed RDS. The palmitic acid content of lecithin from tracheal washing correlated well with the development or absence of RDS and might be used as another parameter in the management of RDS.
When the value of 70 umole/1 of palmitic acid in amniotic fluid was applied to 15 insulin dependent diabetic pregnancies as the criteria for determing fetal lung maturity, Hood et al (1977) found that among eight of those who had values greater than 70 umole/1, one infant developed
RDS; while among seven who had value of 70 umole/1 or less, one had de veloped RDS. They also reported that seven of their 15 diabetic patients had falling level of amniotic fluid palmitic acid at some time before delivery and the decrease could not be explained by change in amniotic fluid volume.
Zuspan et al (1975) suggested the use of palmitic acid/stearic acid
(P/S) ratio of lecithin present in amniotic fluid by gas chromatography as a means of assessing fetal lung maturity.
O'Neil et al (1978) measured the amniotic fluid P/S ratio in 31
insulin-dependent diabetic and 33 non-diabetic patients within 72 hours of delivery. Seven of the 33 infants of nondiabetic mothers
developed RDS while 8 of the 31 infants of diabetic mothers developed
RDS. When each of the groups (diabetic versus non-diabetic) were
considered separately, the mean P/S values distinguished RDS from
non-RDS patients (p< 0.001). For diabetic patients, a P/S ratio of 10 26 was not associated with RDS, while seven out of 23 infants with P/S
ratio below 10.0 did not develop RDS. For non-diabetic patients, a
P/S ratio of greater than 6.0 was not assciated with RDS while 6 out
of 26 infants with P/S ratio less than 6.0 did not developed RDS.
However, the mean values of P/S ratios in diabetic and non-diabetic
groups without RDS were not significantly different. For those who
developed RDS, the P/S ratios of diabetic group were higher than the
non-diabetic group. The data suggested that while the amount of amnio
tic fluid palmityl lecithin is good indicator of RDS, it was not abso
lute, and some other factors were probably involved in fetal lung
maturity. Phosphatidyl glycerol might be one of the factors (Kulovich
et al 1979). O'Neil et al mentioned that if P/S ratio is to be useful
for predicting fetal lung maturity, it is essential that the amniotic
fluid to be free from contamination by blood. A one-percent contamina
tion of the amniotic fluid (v/v) by blood changes the P/S ratio of
amniotic fluid by 50%.
GC-MS Profiling of Organic Acids
While many publications deal with amniotic fluid fatty acids in
relation to fetal lung maturity, or other specific organic acids, only
a few reports have profiled total organic acids in amniotic fluid.
Such profiling requires the availability of expensive instrumentation.
In 1972 Hagenfeldt and Hagenfeldt first reported a gas chromato-
graphic-mass spectrometric analysis of organic acids in amniotic fluid
obtained from 18 normal women in 15-20 weeks of gestation. Utilizing
a LKB 9000 GLCMS system and a coiled glass column (length 2 m, I.D.
4mm) packed with 4% OV-225 on acid-washed and silanized Chromosorb W 27 (100-120 mesh), they quantitated the following organic acids (mmole/1): lactic 6.7 + 2.9 ; pyruvic 0.13 + 0.04; 3-hydroxybutyric 0.46 + 0.021; succinic 0.009 + 0.007; 2-oxoglutaric 0.013 + 0.008; and citric
0.26 + 0.13. By pooling five samples together, they also identified the following organic acids: 2-oxoisovaleric, 2-oxoisocaproic, malic, p-hydroxybenzoic, lauric, myristic, palmitic, stearic and oleic.
Salicylic acid was also found and was attributed to exogenous origin.
Stafford et al (1976) reported a capillary gas chromatographic profile of volatile components of amniotic fluid which was collected 4 h before delivery. They used only 500 ul of amniotic fluid which was extracted with ammonium carbonate-diethyl ether as a salt-solvent pair.
However, the concentration of the volatiles were so low that mass spectrometric analysis was not possible.
Nicholls et al (1978) identified about 30 acids from 40 assumed normal samples of amniotic fluid. These fluid samples fell into two main gestational groups, 15-20 weeks and 30 weeks-term. Their sample hand ling involved centrifugation to remove cells, extraction with ethyl acetate after salting out, pH adjustment, removal of neutral and basic substances, and derivatization. Their instrumentation included a Varian
MAT 311 mass spectrometer interfaced with a Varian 2740 gas chroma tograph using a Watson-Bieman helium separator; and Varian 100 MS data system. Silylated samples were separated on 3% SE-30 on Chromosorb W,
AW-DMCS in a 1.80 m x 2 mm i.d. glass column. They identified 36 compounds which included organic acids, sugars and neutral substances.
Retention relative to the internal standard 2-hydroxyphenylacetic acid and key mass spectral ions (m/z) were given. An alternative procedure utilizing an oxime derivatization prior to extraction was 28 used to help the identification of 2-keto-acids such as pyruvic,
2-ketoisovaleric, 2-ketoisocaproic, 2-keto-3-methylvaleric,
2-ketocaproic and 2-ketoglutaric. A single typical profile was presented.
Wilkinson et al (1979) in a continuation of the paper mentioned above (Nicholls et al 1978), quantitated 13 organic acids from 56 amniotic fluid samples, 35 of which were classified as early gestation and 21 as late gestation. The acids which were quantitated included
2-hydroxybutyric, 3-hydroxybutyric and 3-hydroxyisobutyric, 3-hydroxy- isovaleric, succinic, glyceric, 2-methyl-3-hydroxybutyric, malic, homovanillic, citric, 4-hydroxyphenyllactic, palmitic, and stearic.
When the concentrations of the organic acids of the early gestational group were compared with that of the late gestational group, there was a general decline of amniotic fluid organic acids with advancement of pregnancy.
Shinka et al (1978, 1979) analyzed 24 samples of amniotic fluid from women in 20-40 weeks gestation. Their treatment of the samples included deproteinization with cold ethanol, acidification, sequential extraction with diethyl ether and ethyl acetate, and derivatization with
N,0-bistrimethylsilyl-trifluoroacetamide (BSTFA). The amino acid fraction was separated with cation exchange resin (AG 50 W x 12) after
deproteinization and then trimethylsilylated with BSTFA and acetonitrile.
Analysis was accomplished with a 2 m glass column packed with 3% OV-17,
in a JEOL JMS-D100 mass spectrometer coupled with JMA 2000 mass data
system. Nine amino acids were detected. They were valine, glycine,
proline, threonine, glutamic, phenylalanine, ornithine, lysine and 29 tyrosine. In addition they identified 25 compounds in 46 peaks in their gas chromatogram of the organic acid fraction. Nine compounds were quantitated as a function of gestation period. Lactate and glycolate decreased sharply with gestation. Stearic, palmitic and 2-hydroxy- butyric appeared to decrease from 20 weeks to 34 weeks and then remained relatively constant. Citric, succinic, oxalic and 3-hydroxybutyric acid reached maximum at 28-34 week gestation and then decreased sharply.
Williams et al (1979) utilized an adsorption method instead of the usual solvent extraction or ion exchange methods. Their procedure in cluded methoxime derivatization of keto-acids in the amniotic fluid, adsortion of the derivatized sample onto silica gel, elution of organic acids from the silica gel with tert-amyl alcohol/chloroform/diethyl ether solvent system, drying, trimethylsiliating with BSTFA, and analy sis with gas chromatograph/ mass spectrometer (GC/MS) system consisting of DPD 11/40 computer, Varian 2700 GC coupled to Associated Electrical
Industries MS9-02 mass spectrometer with a silicone membrane molecular separator). Separation was done on glass columns (182 cm x 2 mm i.d.) containing 10% 0V-17 on Chromosorb Q 80-100 mesh. A comparison of the adsorption method with the usual extraction procedure showed that the adsorption method gave a much higher recovery of organic acids from either amniotic fluid or urine samples. The various organic acids identified by Williams et al, Nichollas et al, and Shinka et al are summarized in Table 3. 30
Table 3. Organic acids identified by Williams et al 1979, Nicholls et al 1978, and Shinka et al 1979.
Organic Acids Williams et al Nicholls et al Shinka et al Wilkinson et al ______glycolic + t + lactic + + + pyruvic + + 2-hydroxybutyric + + + 3-hydroxybutyric + + + 3-hydroxyisobutyric + 3-hydroxyisovaleric + 3-hydroxypropionic + cis-glyceraldehyde + trans-glyceraldehyde + oxalic + urea + + 2-ketocaproic + 2-ketoisocaproic + dihydroxyacetone + phosphoric + + + succinic + + + tartronic + maleic + glyceric + + 2-methy1-3-hydroxy- + butyric fumaric + 2-deoxytetronic + lactic dimer + malic + + 2-hydroxybenzoic + 2-hydroxyphenylacetic + 2-hydroxyglutaric + + 3-hydroxy-3-methyl- + glutaric 4-hydroxybenzoic + 3-hydroxyphenylacetic + 4-hydroxyphenylacetic + gentisic + pyroglutamic + lauric + + alpha-glycerophosphate + beta-glycerophosphate + 2-ketoglutaric + homovanillic +
t '+' indicates the particular compound was reported by the corresponding authors. 31 Table 3. Organic acids identified (continued)
Organic Acids Williams et al Nicholls et al Shinka et al Wilkinson et al ______
citric + + + myristic + + vanillic + isocitric + fucono-1,4-lactone + 4-hydroxyphenyllactic + + glucuronolactone palmitic + + hippuric
heptadecanoic + + + + + dodecanedioic + uric + oleic + + linoleic + stearic + + + arachidonic + 2-keto~3-methylvaleric + 2-ketoglutaric + 32 Zambotti et al (1975) identified a few organic acids from amniotic
fluid. These included 4-hydroxy-3-methoxymandelic acid (VMA), 4- hydroxy-3methoxyphenyylacetic acid (HVA), p-hydroxyphenylacetic acid
(p-HPAA), p-hydroxyphenyllactic acid (p-HPLA) and hippuric acid.
Sample treatment involved salting, extraction with ethyl acetate, derivatization to give pentafluoropropionyl methyl esters of the phenolic acids, gas chromatographic and mass spectrometric analysis using LKB
9000-S instrument equipped with LKB 9060 multiple ion detector peak matcher unit with 300 cm x 0.3 cm column packed with 3% SE-52. Because
of low concentrations of HVA and VMA in amniotic fluid, they were
identified by mass fragmentography. There appeared a downward trend
in the concentration of p-HPLA with gestation. One amniotic fluid
sample was analyzed to give approximate values of p-HPAA (100 ng/ml),
HVA (250 ng/ml) and VMA (60 ng/ml).
Muskiet et al (1978) used 3-trideuteromethoxy-4-hydroxy-phenylman-
delic acid (VMA-d3) and 2,5,6-trideutero-homovanillic acid (HVA-d^) as
internal standards and quantitated VMA and HVA in amniotic fluid obtained
from the 15-17th and 32-40th week of gestation. The samples were salted,
extracted with ethyl acetate, derivatized with diazomethane (in diethyl
ether), and again derivatized with pentafluropropionic anhydride and
analyzed with a 2 ra coiled glass column (2 mm i.d.) packed with either
3% 0V-1 or OV-225 on Supelcoport, using a Varian Aerograph 1400 gas
chromatograph coupled to a Varian MAT 112 mass spectrometer equipped with a slit separator and a four channel selected ion-monitoring device.
The concentration of HVA & VMA during 15-17th week of gestation were
20.4 + 6.8 ug/1 from 25 samples and 7.1 + 1.7 ug/1 from 26 samples,
respectively. The values of HVA and VMA in amniotic fluid were compared 33 graphically to those in urine from newborns. They suggested that HVA and VMA in amniotic fluid were almost entirely of fetal origin. Muskiet et al also identified p-hydroxyphenyllactic, hipuric and 5-hydroxyindole acetic acids in the amniotic fluid. They believed that the determina tion of catecholamine metabolites in amniotic fluid could be used for diagnosis of congenital neuroblastoma and maternal pheochromocytoma.
5-hydoxyindoleacetic Acid
The concentrations of 5-hydroxindoleacetic acid (5-HIAA) in amniotic fluid have been measured. It was demonstrated that amniotic fluid con tained high monamine oxidase (MAO) acitivity which metabolized serotonin
(5-hydroxytryptamine, 5HT) into 5-HIAA. Brzezinski et al (1962) believed this prevented the passage of 5HT from the mother to the fetus. Sero tonin is the most powerful known constrictor of umbilical artery. By means of thin layer chromatographic method, Loose and Paterson (1966) identified 5-HIAA in amniotic fluid and measured the concentration of
5-HIAA in 115 samples of amniotic fluid using a spectrophotofluori- metric method. Their results showed that 5-HIAA was not detected in amniotic fluid during the first half of pregnancy but appeared in the second half, showing an increasing tendency with the maturity of the fetus. Samples from toxemic pregnancies had significantly higher concentration of 5-HIAA. The concentration of 5-HIAA from non-toxemic patients during the gestational periods of 36-37, 38-39, 40-41, and
42—43 weeks were 45.5 +5.0, 39.5 + 14.8, 82.2+58.4, and 101.3 +
65.3 ng/ml respectively. The values of 5-HIAA for toxemic patients for the same gestational periods were 60.7 + 55.5, 113.0 +72.4,
129.5 + 72.2 and 193.7 + 58.6 respectively. Anton and Sayre (1971) reported the mean value of 5-HIAA in 3 samples of amniotic fluid to be 34 140 ng/ml which was within the range of values of Loose and Paterson.
Emery et al (1972) compared the 5-HIAA contents in amniotic fluid of normal pregnancies with those from pregnancies in which the fetus had neural tube defects (anencephaly, spina bifida or anencephaly with spina bifida). They found that the latter had lower 5—HIAA in amniotic fluid.
They suggested that the reduced level might be attributed to the reduction of activity of brain tissue because of its malformation. It should be pointed out that Emery et al expressed their measurement in ng of
5-HIAA/mg of protein. It is known that amniotic fluid from pregnancies complicated with neural tube defects contains a higher concentration of protein than that from normal pregnancies. Hence, it is not clear how much this factor contributed to their measurements. Tu and Wong (1976) compared 8 cases of toxemic pregnancy with 18 cases of normal pregnancy.
Although the amniotic 5-HIAA in the former group was slightly higher than that of the latter, there was no statistical significance between
them. The mean values of 5-HIAA in amniotic fluid from the normal group and the toxemic group were 99 +10 ng/ml and 112+23 ng/ml respectively.
By inactivating MAO activity in amniotic fluid, Jones and Pycock (1978) measured both 5HT and 5-HIAA before and during labor. They found that amniotic fluid at labor contained a higher amount of 5-HIAA and 5HT and suggested that the increased fetal excretion of 5HT into the amniotic cavity during labor might induce or enhance uterine contraction.
Citric Acid
The concentration of citric acid in 82 samples of amniotic fluid were measured by Anteby et al (1973). During the 28-43rd week of gesta
tion, there was a statistically significant gradual decline in the mean
levels of citric acid. The values in mg/100 ml of amniotic fluid for the gestational periods (weeks) 19-24, 28-32,33-36, 37-40 and 41-43 were 6.62 + 0.41, 6.68 + 0.51, 5.32 + 0.35, 3.92 + 0.29, and
2.79 + 0.14 respectively. It was suggested that the decline in amniotic citric acid levels might be related to incresed formation and ossification of fetal bone. Shinka et al showed that amniotic citric acid increased from the 20th week, peaked at 31-34th week, and fell sharply to half of its peak value at 39-40 week. Other workers have also identified citric acid in their organic acid profiles of amniotic fluid (Hagenfeldt and Hagenfeldt, 1972; Williams et al, 1979; Nicholls et al, 1978; Wilkinson et al, 1979). The decline of citric acid in amniotic fluid was also observed in fetal guinea pig (Fenton and Nixon,
1974).
Lactic Acid
Lactic acid is one of the most abundant organic acids in amniotic fluid as shown from published profiles of amniotic organic acids
(William et al, 1979; Shinka et al, 1979; Nicholls et al, 1976, 1978).
It amounted to 6.7 +2.9 mmole/1 at 13-20 weeks of gestation (Hagen feldt and Hagenfeldt, 1972). Hendricks (1957) measured the concentra
tion of lactic acid in 52 samples of amniotic fluid ranging from 32-43 weeks' of gestation. The range was 52.4-160.0 mg/100 ml averaging
77.9 mg/100 ml. He found no significant variation in lactic acid within this range of gestational period. However, Shinka et al (1979) showed a rapid decrease in the concentration of amniotic lactic acid
from the 20th to 40th week of gestation. Hendricks also observed a descending gradient of concentrations of lactic acid from amniotic
fluid, through umbilical artery, umbilical vein, uterine vein and maternal arm vein. It appeared that the fetal kidney contributed to 36 the high level of lactic acid in amniotic fluid and this implied that the fetus depended in part upon glycolysis for part of its oxidation energy.
Seeds et al (1979) found insignificant difference in the mean amniotic lactic acid level betwen 56 metabolically well-controlled diabetic patients (9.3 +3.0 mEq/1) and 33 non-diabetic patients (8.6
+ 2.3 mEq/1).
D-(-)-beta-hydroxybutyric Acid
In starving mammals or people with diabetes mellitus, there is a high level of acetoacetate and beta-hydroxybutyrate in blood and urine.
Smith and Scanlon (1973) investigated the concentration of D-(-)-beta- hydroxybutyric acid (BOH) in the amniotic fluid from 53 normal pregnan cies, and 22 pregnancies which (subsequently) terminated in the delivery of dysmature infants. BOH was assayed by an enzymatic method using bacterial beta-hydroxybutyrate dehydrogenase and measuring the forma
tion of stochiometric equivalents of reduced nicotinamide adenine dinucleotide. Results showed that the concentrations of BOH in amniotic fluid and maternal venous blood were not directly correlated. When they examined pregnancies of more than 30 weeks of gestation, they found that the mean amniotic BOH level was significantly higher in immature (0.21 mM) than in normal pregnancies (0.11 mM, p<0.001).
This higher level of amniotic BOH in dysmature pregnancies could result from placental insufficiency and the consequent utilization of fat by
the fetus. Nicholls et al (1976) reported the concentration of BOH
from 22 samples to be 0.12 mM with a range of 0.06-0.24 mM. Seeds et al measured BOH in 61 samples of amniotic fluid from the third tri mester diabetic pregnancies and 25 samples from nondiabetic pregnancies. The results were 0.22 +0.38 mM and 0.19 + 0.23 mM, respectively.
The amniotic BOH level from diabetic pregnancies appeared slightly higher than that from the nondiabetic group, but statistically insigni
ficant. However, there was a positive correlation between amniotic
BOH level and fasting maternal plasma glucose level (p<0.02). The authors believed that this was an indication of placental transfer of
BOH from the maternal side. The insignificant difference in levels of
BOH in amniotic fluid between diabetic and nondiabetic groups was
probably a result of good metabolic control of the diabetic pregnancies.
BOH was also identified by gas chromatographicmass spectrometric methods (William et al, 1979; Nicholls et at, 1978).
2-Hydroxybutyric Acid
Nicholls et al (1976) quantitated 2-hydroxybutyrate in 22 samples
of amniotic fluid. They found the mean concentration to be 0.05mM.
Methylmalonic Acid
Increased concentrations of methylmalonic acid (MMA) were found in
the amniotic fluid of pregnancies at risk of methylmalonic acidemia
(Morrow et al, 1970; Mahoney et al, 1973; Gompertz et al, 1974; Ampola
et al, 1975; Morrow et al, 1977). A rapid method for the analysis of
urinary methylmalonic acid was described by Gibbs et al (1972). The
procedure included extraction with diethyl ether, trimethylsilylation
and gas chromatography. Gompertz et al (1974) reported that the
concentration of MMA in the amniotic fluid from a 17th week pregnancy was 9 ug/ml while the normal range was 0-0.1 ug/ml. This patient had
two previous pregnancies and delivered two infants who suffered severe
metabolic acidosis because of the accumulation of abnormal level of MMA 38 in blood and urine. One died and the other suffered retarded mental and physical development. The abnormality in this case was due to the absence of methylmalonyl coenzyme A mutase activity and was unresponsive to vitamin B-12 treatment. There was abnormal excretion of MMA in maternal urine at 35th week (33 mg/day compared to normal levels of less than 9 mg/day).
Ampola et al (1975) reported a case of successful prenatal diagnosis and prenatal treatment of methylmalonic acidemia. The patient had previously delivered a child who died of methylmalonic acidemia at 3 months of age. The disease was diagnosed posthumously. During the second pregnancy, amniotic fluid cells were cultured at the 19th week of gestation. The fetus was diagnosed to be defective in the synthesis of 5 '-deoxyadenosylcobalamin and would suffer methylmalonic acidemia.
The amniotic fluid showed an increase level of MMA. At the 19th week, the amniotic fluid contained 2.19 ug/ml while 6 controls at 16-24 weeks' gestation averaged 0.76 + 0.38 ug/ml. The mother was subsequently treated during the last nine months of pregnancy with a large dose of vitamin B-12 which decreased maternal urinary excretion of MMA. At term the amniotic fluid contained 4.71 ug/ml of MMA while 7 controls of 27-38 weeks' gestation average 1.25 + 0.54 ug/ml. The baby was delivered in excellent condition. The first voided urine contained 67 ug of MMA/mg of creatinine, while 6 normal newborns gave a value of
2.1 + 0.7 ug/mg. The baby grew normally under a restricted diet.
A rapid method for prenatal diagnosis of methylmalonic acidemia was reported by Morrow et al (1977). The procedure consisted of measuring the concentration of MMA in amniotic fluid and the activity of methylma lonyl coenzyme A mutase in uncultured cells of amniotic fluid. MMA levels 39 in amniotic fluid were measured by extraction with ether and analysed by gas chromatography of the propanol ester derivatives of organic acids.
Propionic Acid
Propionic acidemia is a genetic disease inherited as an autosomal recessive trait characterized by a deficiency of propionyl coenzyme A carboxylase with resultant propionic acidemia, hyperglycinemia and hyper ammonemia. The disease is most commonly lethal very early in life
(Shenfai et al, 1974). Propionic acid is derived from the hydrolysis of propionyl coenzyme A, which in term results from the metabolism of va line, isoleucine, methionine, threonine, odd-chain fatty acids, the side chain of cholesterol and propionic acid by intestinal bacteria.
Sweetman et al (1979) reported the detection of two diastereomers of methylcitric acid in the amniotic fluid of a pregnancy the fetus of which was confirmed as homozygous for propionic acidemia. The amniotic fluids of two subsequent pregnancies at risk for propionic acidemia were examined and did not contain methylcitric acid. One fetus was hetero zygous and the other homozygous normal. The acid was also not found in samples of amniotic fluid from normal pregnancies. The concentration of propionic acid in the amniotic fluid of the fetus homozygous of propionic acidemia was 19.57 uM while in 5 normal amniotic fluid the mean level was 14.56 + 2.64 uM. The authors suggested that measurement of methylcitrate might be useful in prenatal diagnosis of propionic acidemia. 40
Bi.le Acids
Deleze et al (1977) measured the concentration of total bile acids in the amniotic fluid from 11 normal pregnancies with gestation period of 28-42 weeks and 9 samples from pregnancies of 28-38 weeks of gesta tion and complicated with polyhydramnios. Two of the latter group delivered infants with intestinal obstruction distal to the papilla of Vater, a condition known to cause regurgitation of bile into the amniotic fluid. The amniotic fluid samples underwent solvolysis, alkaline hydrolysis, extraction with ethyl acetate, methylation, trimethylsilylation, and gas chromatography. Cholic, chenodeoxycholic,
3-betahydroxy-5-cholenoic and deoxycholic acid accounted for 90% of total bile acids in all samples, and the first three bile acids were found in all amniotic fluid samples. The total amniotic bile acid concentration from normal pregnancies ranged from 1.4 to 2.4 uM. For the polyhydramniotic group without fetal intestinal obstruction distal to the papilla of Vater, the values ranged from 0.9 to 1.9 uM.
However, for two of the polyhydramniotic group who had that type of fetal intestinal obstruction, the amniotic bile acid concentrations were 30.3 and 83.1 uM. The authors pointed out that the measurement of bile acid concentration in amniotic fluid could be used for prenatal diagnosis of intestinal obstruction distal to the papilla of Vater.
In a subsequent paper, Deleze et al (1978) reported the bile acid pattern in human amniotic fluid by measuring indivdual bile acids in 29 samples from 26 pregnancies within the gestational period of 32-41 weeks.
The mean total bile acid concentration was 1.57 uM ranging from 0.4-4.8 uM, and showed no significant variation with gestational age(32-41 weeks). 41 Cholic, chenodeoxycholic and 3-beta-hydroxy-5-cholenoic acids were found in all samples, while lithocholic acid was found in ten and deoxycholic aicd in nine samples. The identification of the methyl ester trimethyl- silyl ether derivatives of the bile acids was performed with combined capillary gas chromatography-mass spectrometry (GC-MS) (Finnigan
Model 1015 D with Finnigan Model 6000 Interactive Data System). The gas chromatograph was equipped with a Grop-type injector and a glass
capillary column (Silar 10, precoated with barium carbonate, 25 m x
0.24 mm). Hyodeoxycholic acid O-^-b-^-dihydroxy-Stf-cholanoic acid) was used as internal standard. When the molar percentages of individual
bile acids were compared, 3 hydroxy-5-cholenoic acid averaged 39.8%
of total bile acids during 32-37 weeks of gestation and 20.2% at term
(p<0.01). However, the percentage of cholic acid was significantly
lower (34.0%, p<0.05) during 32-37 weeks of gestation than at term
(55.4%). Deleze et al suggested that the decrease in 3(^-hydroxy-5-
cholenoic acid near term might reflect maturation of hepatic bile
acid biosynthesis.
Employing radioimmunologic techniques, Heikkinen et al (1980)
measured the concentration of cholic, chenodeoxycholic and deoxycholic
acids in amniotic fluid from 7 normal pregnancies and 10 complicated
with toxemia, 9 with diabetes mellitus, and 9 suffering maternal intra-
hepatic cholestasis. The gestational ages of the samples were in the
range of 35-37 weeks. They found that the concentrations of amniotic
cholic and chenodeoxycholic acids, which are primary bile acids, were
much higher in pregnancies complicated with maternal intrahepatic
cholestasis than those in normal pregnancies. The concentrations of
these two bile acids were only slightly elevated in toxemia or diabetic 42 pregnancies. It has been reported that increased risk of fetal morta
lity was associated with maternal intrahepatic cholestasis. However,
it is not known how excessive concentrations of bile acids level in
amniotic fluid would affect the well-being of the fetus.
Uric Acid
Uyeno (1919) detected the presence of uric acid in amniotic fluid.
Williams et al (1924) measured the concentration of uric acid in 20
amniotic fluid samples at labor (11 normal pregnancies, 9 abnormal
ones, the nature of abnormality was not reported). The concentration
of uric acid averaged 4.51 mg/dl of fluid (ranged 1.93-7.73 mg/dl) in
normal pregnancy, and 3.81 mg/dl from the abnormal samples (ranged
1.93-8.51 mg/dl). The difference between the two groups was not
statistically significant. Serr et al (1963) compared the concentration
of uric acid in amniotic fluid with those from maternal and fetal
blood. Fourteen cases of early pregnancy (6-16 weeks) and 51 cases
at labor were examined. Results showed that during early pregnancies,
the mean amniotic fluid uric acid content (3.0 + 0.9'lmg/dl) was not
much different from that of maternal blood (3.8 + 1.3 mg/dl).
However, when the samples at term were examined, the concentration of
uric acid in amniotic fluid (6.7 + 1.76 mg/dl) was significantly
higher (p 0.01) than that in fetal blood (4.7 + 1.24 mg/dl) and in
maternal blood (4.3 + 0.86 mg/dl). At term there appeared a gradient
of uric acid concentration from amniotic fluid to fetal blood to
maternal blood. The results were consistent with the concept that in
early pregnancy, the amniotic fluid is a transudate of maternal plasma,
and at late pregnancy amniotic fluid is mainly the product of fetal
kidney. Harrison (1972) measured the levels of uric acid in 100 samples
of amniotic fluid obtained from the gestational age of 29-43 weeks.
The data showed an increase of amniotic fluid uric acid level with advancement of pregnancy. Furthermore, 79% of all the cases had a value of 8.5 mg/dl or greater at 38 weeks or over. However, this was not adequate for predicting fetal age in utero. Krstulovic (1979)
reported a procedure for rapid determination of uric acid levels in amniotic fluid, utilizing a reverse phase high performance liquid
chromatographic (HPLC) system in which chromatographic peaks were monitored spectrophotometrically at 280 nm, fluorometrically with an
excitation wavelength of 285 nm and emission at 320 nm, as well as
electrochemically at a potential of 0.800 V. Identification was achieved by comparing the stopped-flow UV spectra of the sample with
that of authenic compound. The concentration of uric acid in 14 samples
of amniotic fluid obtained during 15-24 weeks of gestation were deter mined and ranged from 0.897 to 4.39 mg/dl of amniotic fluid. If the
HPLC system is equipped with automatic sample injection system, the method would be very rapid, precise and convenient. 44 4. Organic Bases
While there are many publications dealing with drugs, many of which are organic bases and were detected in amniotic fluid, there are
only a few publications on endogenous organic bases in amniotic fluid.
They will be discussed briefly below.
Polyamines
Putrescine, spermidine and spermine are generally referred to as polyamines. They stimulate a large number of processes of replication,
transcription and translation. The amino moiety of the polyamines is protonated in physiological environments and hence is positively charged. The interaction of these positively charged molecules with
the negatively charged phosphate groups of the ribonucleic and the deoxyribonucleic acids must play an important role in the processes of protein, DNA and RNA syntheses.
Using both an amino acid analyzer and a two-dimensional thin layer chromatographic system with the lower limit of sensitivity at 50 pmole per ml, Chan et al (1979) were unable to detect free polyamines in human amniotic fluid obtained from 13th to 40th week of gestation.
However, they did find polyamines conjugated to peptides and proteins
in the amniotic fluid. By using gel filtration and ion-exchange chroma
tography, Searle et al (1979) purified the putrescine-protein conjugate
Ninety to 100 percent of amniotic putrescine was found associated with a protein of approximately 4600 daltons. Spermidine was entirely associated with the 10,000-30,000 dalton fraction while spermine was
identified in the 1000-10,000 dalton fraction as well as in a low molecular weight fraction. The biological significance of polyamines 45 in amniotic fluid, beyond the fact that polyamine are found whenever
there are cell growth and development, is not known.
Catecholamines
Zuspan et al (1974) reported a comparative study on the levels of
epinephrine (E) and norepinephrine (NE) (in ug/mg of creatinine) in
human and baboon amniotic fluids. They found there was no significant
difference in the level of NE (ug/mg of creatinine) in amniotic fluids
between human and baboon. However, baboon amniotic fluid contained
significantly higher levels of epinephrine than that of human. The
ratio of E/NE in human amniotic fluid was 1.23 while that of baboon was 4.34. It was suggested that perhaps this reflects that the adrenal
medulla was more fully developed in the baboon than in the human fetus.
5-hydroxytryptamine (Serotonin)
After inactivation of amniotic fluid monoamine oxidase activity by
protein precipitation, Jones and Pycock (1978) measured the concentra
tion of 5-OH-tryptamine (5HT) in amniotic fluid with a fluorometric method. Their results showed that patients who were in labor had
significantly higher concentration of 5HT in the amniotic fluid
(0.05 + 0.08 ug/ml) than those who were not in labor.
The scarcity of free endogenous amines in the amniotic fluid is
probably due to the activities of monoamine oxidases and diamine
oxadase in amniotic fluid compartment. 46 5. Hormones
Hormones in amniotic fluid have been recently reviewed (Josimovich,
1978; Belisle and Tulchinsky, 1980). Various hormones and their meta bolites found in amniotic fluid are listed in Table 4.
Table 4. Steroidal Hormones and Metabolites in Amniotic Fluid c18 estrone (Ej) C21 estradiol (E2) estriol (E3) 5-beta-pregnan-20-one estetrol (E4) 17-0H-progesterone 1l-dehydro-estradiol-17-alpha progesterone 16-oxo-estradiol pregnanediol-glucuronide 2-methoxyestrone 3-alpha-0H-5-beta- 16-alpha-hydroestrone pregnan-20-one 16-beta-hydroestrone 3-alpha~6-alpha-di-OH- 15-alpha-hydroestrone 16-alpha-OH-progesterone E^-lb-glucuronide pregnenolone E3-3-sulphate-16-glucuronide 17-OH-pregnenolone E3~3-sulphate pregnendiol E3-I6-sulphate 17-0H-pregnenolone-sulphate E3~3-glucuronide cortisol cortisone c19 17-0H-corticosteroids 6-beta-0H-20-alpha-dihydrocortisol dehydroepiandrosterone 6-beta-0H-20-beta-dihydrocortisol dehydroisoandrosterone 6-beta-OH-cortisol androstenedione corticosterone testosterone dehydrocorticosterone testosterone-glucuronide deoxycorticosterone-sulphate testosterone-sulphate tetrahydrocortisol dihydrotestosterone tetrahydrocortisone androstenediol tetrahydrocortisone-glucuronide 16-alpha-OH-dehydroepi- corticosterone-sulphate androsterone cortisone-sulphate 16-ketoandrostenediol cortisol-sulphate androstenetriol 11-OH-corticosteroids 16-OH-dehydroepi- androsterone-sulphate 16-ketoandrostenediol-sulphate Other hormones in amniotic fluid are also listed below in Table 5.
Table 5. Non-steroidal Hormones in Amniotic Fluid.
Prostaglandins: PGEj, PGE2 , PGFa l , PGFa2
Thyroid Hormones:
diiodothyronine, T2; triiodothyronine, T3; tetraiodothyronine, T4; reverse-triiodothyronine, rT3
Peptide Hormones:
human chorionic gonadotropin, hCG; follicle-stimulating hormone, FSH; luteinizing hormone, LH; prolactin; human placental lactogen, hPL; human growth hormone, hGH; insulin; glucagon; human corticotropin, ACTH; oxytocin; neurophysin; beta-MSH; somatomedin; somatomedin C; beta-endorphin 6. Drugs
The placental transfer of drugs from the maternal compartment to
the fetus after administration to the mother has been well established
and documented (Moya and Thorndike, 1963; Villee, 1965; Mirkin, 1973;
Horning et al, 1973). The so-called placental barrier between the
mother and the fetus is actually a relative impedance to the transfer
of drugs and foreign matters between the two compartments. This
barrier is dependent, in part, upon the following characteristics;
lipid solubility of the drug, ionization state at the given pH, mole
cular size, and the concentration gradient. In addition this barrier may be breached by transport molecules or metabolic processes. Pino-
cytosis seems to have little importance in the placental transfer of
drugs though it might be important immunologically. If these placen-
tally transferred drugs survive elimination by the metabolic processes
of the mother, fetus and the placenta, some ends up in the amniotic
fluid. Various drugs administered or chemicals exposed to the mother during her pregnancy were detected in the amniotic fluid. A list of
the drugs and chemicals identified in amniotic fluid are given below
in Table 6.
Table 6. Drugs or Chemicals Detected in the amniotic Fluid. adrenaline barbituric acid derivatives adriamycin betamethasone amikacin biglumide aminophylline bromhexine amoxycillin buphenine ampicillin aromatic amines arterenol Table 6. Drugs or Chemicals Detected (continued).
caffeine meclizine carotene merperidine carotine metampicllin Cathesin D methaqualone cefadroxil methadone cefazolin methicillin cefaprin methyldopa cephacetrile neomycin cephalexin nicotine cephalosporin nitrosahexamethyleneimine cephaloridine nitrosopiperidine cephalothin normerperidine cephradine cinnarizine chloral hydrate & metabolites oxacillin chloramphenicol oxazepan chlorcyclizine oxo-amide chlorpromazine cobalophilin cortisol pentazocine cortisone penicillin cotinine penicillin analogs phenobarbital piromidic acid polychlorobiphenyl Danthron prilocaine DDT properdin DDE pyrocatecholamines dexamethasone dibekacin rifampicin dicloxacillin digoxin SKF 525 A dialkylnitrosamine streptomycin diphenylhydantoin sulfanilamide
epicillin tetracyclines tetrahydrocannabinol gentamicin thalidomide theobromine indomethacin theophylline isocarboxazid thiamphenicol isoniazid thiopental thiourea tobramycin tromethanine
vitamin A vitamin Bl Vitamin B12 vitamin C 50
G. GC-MS Methodology
While the technique of gas-liquid partition chromatography was pioneered by A.T. James and A.J.P. Martin in 1951-1952, the use of selective adsorption for separation of gases and vapors originated before the First World War. The most significant appication was in the gas mask and in commercial processes in the recovery of raw gasoline from natural gas, and benzene and light oils from manufactured gas.
The historical development of gas chromatography was reviewed by Ettre
(1975). Since the publication of the appication of gas-liquid-partition chromatography to the separation of fatty acids by James and Martin in 1952, gas chromatography has been undergoing unprecedented growth.
This is best illustrated by an extensive review on gas chromatography by Cram and Juvet (1975) who covered the development of this field in 1974-1975, citing 1748 references.
The basis of gas chromatography is the distribution of a sample between two phases. The sample in vapor phase is passed through a column containing a stationary phase which can be either a solid, a liquid, or a liquid coating a solid support, or a liquid phase coating the wall of the column. Essentially, different components of the gas eous mixture have different partition coefficients between the gas phase (the carrier gas) and the stationary phase in the column, depend ing on the summation of the physicochemical properties of the sample constituents, the carrier, and the stationary phase. When the station ary phase is a liquid, the separation process is termed gas liquid chromatography (GLC). When the stationary phase is a solid, the method 51 is called gas solid chromatography (GSC). Since most gas chromatographic applications use liquid stationary phase, gas liquid chromatography is commonly referred to as GC.
The basic components of a gas chromatograph consist of a tank of carrier gas, a gas regulator valve for controlling the flow rate, an injection port for introduction of the sample, a column for separating the sample into its individual components, an oven in which the column is placed for temperature programming, a detector and a recorder. A microprocessor or minicomputer can be incorporated for data acqusition, data processing, temperature programming, and even pressure programming.
The following paragraphs will describe briefly the components required for successful GC analyses.
The most commonly used carrier gases are helium, hydrogen and nitrogen, depending on the chromatographic requirements and the detector employed. The carrier gas should be highly purified to reduce back ground, ensure reproducibility and lengthen the lifespan of the column.
Cartridges are available from commercial sources to be incorporated into the gas line to remove moisture, oxygen an impurities before the gas enters the column.
Chromatographic columns can be divided into two main categories, namely, packed columns and capillary columns. Packed columns can be made of glass or metal (e.g. stainless steel). While a metal column
is more sturdy, a glass column has less reactive catalytic sites which sometimes interfere with the chromatographic precess. Packed
columns of 3-16 ft and i.d. of 2-4mm are in common use, relatively
less expensive, easy to handle and very good for routine analysis of
simple mixture or the routine quantitation of particular compound(s). 52
There are three types of capillary column, the wall-coated open tubular
(WCOT) column with the inside wall coated by a liquid phase, the porous- layer open tubular (PLOT) column in which the inner surface has been extended by substances such as fused silica, and support coated open tubular (SCOT) column in which the liquid phase is supported on a sur face covered with some type of solid support material, which increases the surface area for the liquid phase. SCOT columns have higher loading capacity, but are more difficult to make and use. Capillary columns are more expensive than packed columns, much more difficult to use because a small dead volume has a large impact on the resolution of the column, has a smaller loading capacity, can deteriorate very easily but, it gives superb resolution when properly used. Capillary columns of 0.25-
0.50 mm internal diameter and of length of 50-100 m are in common use.
With the number of theoretical plates of 2500-5000 per meter, capillary columns are unmatched in separation of very complex mixture.
Numerous solid support materials and liquid phases are available commercially and they will not be discussed here. The choice of these materials depend on the nature of of the sample to be analysed, in terms of polarity, basicity, acidity, volatility, chemical reactivity and thermal stability. Combined liquid phases are sometimes used for specific applications. The polarity of the liquid phase often deter mines the type of sample that can analysed.
Various detectors are used in gas chromatographs. The most common are flame ionization detectors(FID). Other detectors include thermal conductivity detectors, nitrogen phosphorus detectors, electron capture detectors, flame photometric detectors, microwave emission detectors and electrochemical detectors. These detectors differs in selectivity, 53 sensitivity, specificity, and linear range. The selection depends on experimental requirements. For example, electron capture detector would be best for pesticides anaylsis.
Gas chromatography has been used for both qualitative and quan
titative analyses. Qualitative analysis consists of comparison of the retention time of an unkown to that of an authenic compound. Kovats
retention indices are more reproducible than retention times (Kovats,
1965). Retention times are affected by various parameters, such as flow and temperature fluctuations and ageing of the column. When qualitative analyses are carried by GC with mass spectrometer as detector, the
identification in most incidences is unequivocal, with the exception
of steric isomers. Quantitative analyses by GC are accomplished
by the measurement of peak heights or integration of peak area by
electronic integrators. The quantition of overlapping or partially
resolved peaks remains a challenging problem. Various mathematical methods were proposed to deconvolute unresolved chromatographic peaks.
Mass spectrometry is one of the most specific analytical techni
ques. Its application encompass many disciplines, including physics,
chemistry, cosraoschemistry, geology, atmospheric and enviromental
sciences, biology, medicines, criminology, even history of art and
archeology and many others. A mass spectrometer consists of the
following components, an inlet system, an ionization chamber, a mass
analyzer, an amplifier, a recorder, and a high efficient pumping system.
The sample to be analysed is introduced into the mass spectrometer
through the inlet system, which can be a batch inlet, a direct probe
or an interface coupling a gas chromatograph to the mass spectrometer. At the ion source molecules under high vacuum are bombarded with electrons, lose some outer electrons and become positive ions. Negative ions are also formed but less abundant. In general, positive ions mass spectra are utilized for indentification purposes. The resultant posi
tive ions or ion fragments emerging from the ion source at a given velocity are separted from one another according the to the mass to charge
ratio at the analyser region. For a quadrupole instrument, the analy
zer consists of a quadrant of four parallel hyperbolic or circular
rods which provide a specific oscillating field. At a specified radio
frequency, ions of a given mass undergo stable oscillation between the
electrodes, while ions of lower or higher mass are collected on the
quadrupole rods. The oscillationg ion continues at its original velocity
down the flight path to the collector. A quadrupole instrument has the
advantage of fast scanning, hence suitable for capillary columm gas
chromatographic analyses. However, there is a considerable mass dis
crimination against higher mass, and ions of mass above 250 will have
lower intensities. The quality of the mass spectra obtained from a
quadrupole instrument is highly dependent on the alignment of the poles
and tuning parameters. For a single-focussing magnetic deflection
mass spectrometer, the analyzer region consist of a homogeneous magnetic
field, which separatee the ions according to the mass to charge ratio.
For a high resolution mass spectrometer, the analyzer region contains
both a magnetic sector and an electrostatic sector. The ions eventually
hit the electron multiplier. The signals are amplified and recorded.
The diversity, complexity and rapidity of developments in the area of
mass spectrometry are best illustrated by the excellent reviw by Bur
lingame and Kimble (1976) who covered the a two year period, citing 55 1260 references.
The marriage of the high separative power of capillary gas chroma
tography with the high specificity of mass spectrometry has produced a powerful tool for the analyses of complex mixtures encountered in biomedical and environmental sciences. The coupling of a gas chromato graph to a mass spectrometer, together with the advances of data acquisi
tion and data processing technologies enables multivariate analyses of
the chemical nature of complex biological systems. The application of
GC-MS to the profiling of urine and serum have elucidated many meta
bolic diseases. The extensive application of GC-MS methodology is best
described in the excellent reviews by Roboz (1975), Lawson (1975), and
Jellum (1977). 56 H. Ojective of Present Investigation
The specific objectives of the present investigation are:
(a) To find a suitable procedure for the extraction of organic com
ponents from complex biological mixtures.
(b) To find suitable conditions for the gas chromatographic-mass
spectrometric analysis of the various partitioned fractions from
complex biological samples.
(c) To examine the profile of organic acid components of normal amniotic
fluid.
(d) To examine the profile of organic acid components of amniotic fluid
from disease-state pregnancies.
(e) To examine the profile of basic components of normal amniotic fluid.
(f) To examine the profile of basic components of amniotic fluid from
disease-state pregnancies.
(g) To examine the profiles of the acidic and basic components of
various biological sample related to maternal-fetal medicine.
The ultimate goal is to apply procedures (a-g) above and to find
from amniotic fluid (or other biological specimens) drugs and unique
chemical marker(s) which can be used as definitive diagnostic tools for
prenatal screening for fetal distress and genetic diseases. It is also
hoped that the chemical profiles obtained from the present and future
studies will give more comprehensive pictures of the environment in
which the fetus is suspended during the gestation period. Furthermore,
it is hoped that a better understanding of the overall chemical environ
ment of the fetus may lead to the development of clinical measures to
correct fetal distress and genetic diseases in utero. II. MATERIAL AND METHODS
A. Biological Samples
Amniotic fluid samples were obtained from the Ohio State University
Hospital during amniocenteses or Caesarean sections necessitated for the assessment and management of fetal lung maturity in diabetic or Rh sensitized pregnancies, for prenatal screening of genetic diseases because of maternal age, for the detection of neural tube defects or other anomalies. The samples were usually frozen at -20°C until thawed to be processed. In the instances when samples were processed immedi ately upon collection and without freezing, data collected will be clearly noted.
Other biological specimens were obtained from either the Ohio
State University Hospital or animal facilities at the University.
These included human follicular fluid obtained during hysterectomy, meconium from an infant at birth, urine and blood samples as well as follicular fluids from baboons. These samples were stored at -20° C until use.
All organic solvents used for extractions were of nanograde from
Mallinckrodt, Inc. St. Louis, Missouri or Omnisolv glass distilled from MCB manufacturing Chemists, Inc., Cincinnati, Ohio. Hydrochloric acid solution, concentrated ammonium hydroxide, and saturated sodium borate buffer, which were used for adjusting pH and salting purposes, were extracted 3 times each with equal volume of dichloromethane and
57 ethyl acetate. Trimethylsilylating agent, N,0-Bis-(Trimethylsilyl)tri- fluoroacetamide (BSTFA) was obtained from Pierce Chemical Co., Rockford,
Illinois. Deutero-Regisil from Regis Chemical Co., Morton Grove,
Illinois. Authentic organic acids were obtained from Sigma Chemical
Co., St. Louis, Missouri ; Analabs, Inc., North Haven, Connecticut, and Applied Science, State College, Pennsylvania.
B. Extraction of Amniotic Fluid
Initially amniotic fluid samples were extracted with chlorofrom/ isopropanol (3:1) solvent system according to the procedures outlined in the Extraction Flow Chart 1 (see Figure 1). Subsequently, the extraction method was modified as detailed below.
Most of the AF samples came in glass syringes which were placed in freezer at -20° C. In order to minimize enzymatic reactions during the thawing process, a 10 ml graduated cylinder was placed in ice while the syringe, with the needle removed, was supported above the graduated cylinder such that the thawed portion of the amniotic fluid flowed into the ice-cold graduated cylinder. Internal standards, heptadecanoic acid and phenyl-d5-mandelic acid were added to the sample each at a concen tration of 1.5 ug/ml. The sample was mixed and transfered to a pyrex glass tube and centrifuged at 10,000 x g for 10 minutes using Sorvall centrifuge with rotor ss-34. After centrifugation, the supernatant was gently poured into a separatory funnel with a glass stopcock and a glass stopper. To every 5 ml of amniotic fluid, 10 ml of pre-extracted saturated sodium borate buffer was added both for salting effect and for adjusting the pH to a value of 8.5-9.0. Then 20 ml of dichloro methane was added for every 5 ml of amniotic fluid. The separatory 59
Amniotic Fluid (AF, 5-7 ml)
a. add 10 ml saturated ^26407 soln. to adjust pH 8.5
b. extract 3x 10 ml CHC^/isopropanol (3:1), by vortex, centrifuge I Organic layer (.0 ml) Aqueous layer (17 ml)
(neutral, bases, phenols, alcohols) I a. add 5 ml 1 N H2SO4 a. + 10 ml H2SO4 vortex, sep upper layer b. ext. 3x CHCI3/ b. repeat 2x more isop. 5 ml,vortex
Organic ulous (15 ml) Aa r ~ (alcohol & neu) (bases) Organic acids Aqueous (discard)
evap. dry a. + 3 ml conc. evap. dry i NH4OH pH 9-14 i GC-MS b. ext. 3x CHCI3/ ch2n2 f isop., 5 ml I neutral methyl est. f L Aqueous TOrganic — * • dry (discard) (bases) i I TMS dry I * TMS GC-MS t GC-MS
Figure 1. Extraction Flow Chart 1. 60
funnel was hand-shaken vigorously for 15 seconds and the emulsion-like
content was drained into pyrex tubes which were centrifuged at 2500
rpm for 10 mins. with an IEC model K centrifuge (International equip
ment Co., Needham Hts., Mass.). The upper aqueous layer was trans
ferred with a pasteur pipette into the separatory funnel to be extrac
ted two more times with dichloromethane. The lower organic layer
usually contained white, fluffy, presumably proteinaceous material which could be removed from the organic phase by stirring it with a
pasteur pipette and rotating the tube while decanting the organic
solution. The rotation made the fluffy material stick to the side of
the tube, thus facilitating the removal of the organic phase. The
dichloromethane portions were pooled and labelled as Fraction A while
the aqueous portion was Fraction B (see Extraction Flow Chart 2). The
volume of Fraction A was reduced to about 10 ml by means of a rotary
evaporator connected to an aspirator and the water bath temperature
at 35-40° C. Fraction A, containing both neutral and basic components, was further partitioned into Fraction C (neutrals) and Fraction H
(bases) as follows.
Fraction A was placed in a separatory funnel and washed 3 times
each with 5 ml of 0.5 N HC1. The bases were protonated and became
soluble in the aqueous phase while the neutrals remained in the organ
ic phase. The volume of the washed organic phase was reduced to about
1 ml with a rotary evaporator and transferred to a small vial (1 dram
size, with a screw-cap) and then blown dry with purified nitrogen at
35-40°C. This was labelled as Fraction C. Until analysis by gas
chromatography and mass spectrometry, all dried samples were stored at Amniotic Fluid (5 ml)
7.5 ug d5~Mandelic acid 7.5 ug heptadecanoic acid centrifuge 10,000 x g; 10 mins. 4°C
Supernatant Pellet (discard)
10 ml sat. sodium borate buffer pH (8.5-9) separatory funnel extract with 3x20ml CH2CI2 (15 sec.) centrifuge 2500 rpm . 10 mins at 25°C
Pooled Orgoiiju>. v^/ Aqueous (B)
washed 3 x 5ml 0.5 N HC1 1 ml 3N HC1 pH (1.0) Organic (C) Pooled Aqueous (D) extract 3x20ml (neutrals) I I CH2Cl2 2ml conc. n h 4oh Pooled Org.(El) Aqueous (Fl) 2x20 ml CH2CI2 2x20ml ETOAC
Pooled 0rg.(E2) Organic Aqueous(G) (.bases) (discard) Aqueous (F2) heat 100°C (H) 15 mins cooled 3x20ml CH2CI2
Pooled Org.(E3) Aqueous (F3) 2x20ml ETOAC
Pooled Org. (E4) J Aqueous (F4) (discard)
Figure 2. Extraction Flow Chart 2 -20°C in vials of one-dram size and sealed with screw-caps lined with aluminum foil.
To the aqueous phase containing the bases, 2 ml of concentrated ammonium hydroxide was added. The pH was checked (11-12) and the solution was then extracted 2 x 20 ml of CH2CI2. The pooled organic phase was dried as described above, labelled as Fraction H and properly stored.
The pH of the aqueous fraction (B) was adjusted to 1.0 by adding
1 ml of 3N HC1. It was then extracted with 3 x 20 ml of CH2C12* Each extraction was followed by centrifugation at 2500 rpm for 10 mins. at
25°C. The pooled organic phase was labelled as Fraction El and the aqueous phase as Fraction FI which was subsequently extracted with 2 x
20 ml of ETOAC. The organic phase was pooled as E2 and the aqueous fraction (F2) was placed in an erlemyer flask which in turn was placed in a beaker with boiling water. When F2 started boiling inside the flask, it was left there for 15 more minutes. The flask was then removed from the boiling water bath and cooled under tap water. After cooling, fraction F2 was extracted again with 3 x 20 ml of CH2CI2 and the organic fraction was pooled as E3 which contained acid hydro lysed acids. The aqueous fraction F3 was further extracted with 2 x
20 ml of ETOAC. The organic fraction was pooled as E4 and the aqueous fraction was discarded. Each of the organic fractions was dried and stored in a small screwcapped vial as described above.
C. Derivatization of Acidic Fractions
The procedure for trimethylsilylating acidic fractions was as follows. A previously prepared sample was removed from storage from a freezer, warmed up to room temperature to avoid condensation of moisture, and placed on a heating block at 70° C for one minute. Then the screw cap loosened and 1 ul of triethylamine was added to the vial, followed by 30 ul of N,0-Bis-(Trimethylsilyl)-Triflouroacetamide, and the cap was immediately screwed tight. The vial was heated at
70° C for 35 minutes. The sample was allowed to cool and now was ready for gas chromatographic-mass spectrometric analysis.
Some acidic samples first underwent methyl esterification before trimethylsilation. To a sample which had been warmed up to room tem perature, 1 ml of ether was added to dissolve the acids. Diazomethane was bubbled through the etheral solution until it appeared yellow, indicating excess of diazomethane in the solution. Then the diazo methane generator was removed and the solution containing methyl esters was blown dry by nitrogen gas at 35-40°C. The sample was next tri- methylsilylated as described above. The generation of diazomethane is described below.
A simple design for generating diazomethane is shown in the fol lowing diagram (Fig. 3). It consists of a 50 ml erlemyer flask, a newly designed (Patent Application Pending) connecting glass tube with one end of a diameter close to that of the mouth of the erlemyer flask and the other end drawn out to a diameter of about 4 mm for bubbling diazomethane into the etheral (or dichlororaethane) solution of acids to be methylated. To 20 ml of H2O in a 50 ml erlemyer flask, 15 KOH pellets were added, and the flask was swirled to dissolve the pellet. Then 10 ml of MeOH was added to the KOH solution, followed by 15 ml of diethyl ether. One hundred to 300 mg of Diazald (N-Methyl
-N-nitrosop-toluenesulfonamide) on the tip of a spatula as well as a small teflon coated magnetic stirrer were added to the flask. 64
glass tube
teflon collar
50-ir.l Erlemver flask
Stirring bar (teflon)
sample Stirer & Heater
Fig. 3 Diazomethane micro-generator 65
The gas collecting tube was tightly fitted to the mouth of the flask with a short collar made from a teflon tubing. The flask was gently heated on a hot plate with continuous stirring. As soon as diazo- methame, a light yellow gas, began to bubble from the solution, the rate of formation could be controlled by adjusting the rate of stirring and heating. The gas was bubbled through the solution to be derivatized.
As diazomethane reacted with the acids, the yellow color disappeared.
After all the acids were derivatized, the solution turned yellow indicating excess of diazomethane and the completion of the derivatizing process. The derivatized sample was gently dried in a stream of dry nitrogen gas at 35-40°C, and then further derivatized by trimethylsila- ting agent as described above.
D. Gas Chromatographic- Mass Spectrometric Analysis of Samples
Samples were analysed with either a low resolution quadrupole mass spectrometer and/or a double focussing high resolution magnetic sector mass spectrometer. The low resolution mass spectrometer (Hewlett
Packard 5840A GC/MS system) consisted of a quadrupole mass spectrometer, a 5940A HP gas chromatograph, a 21MX E-series computer, a Tektronix
4012 terminal, a Tektronix 4631 hard copy unit, a HP 7970B digital tape unit and a Zeta plotter (100 series). The high resolution mass spectrometer system consists of a Varian MAT 311A, a Varian gas chroma tograph Model 3700, a Varian Spectro System MAT 200 with a Digital PDP
11/34 computer, a Pertec tape unit, a Tektronix 4010-1 terminal, and a
Statos 41 printer/ plotter.
In order to perform a GC/MS analysis, a suitable gas chromato graphic column was needed to separate a complex mixture into the iso lated components. The selection of a proper column depends on the nature of the mixture to be analysed, the type of separation required, the solid support needed, the liquid phase coating the solid support or coating the interior wall of the glass column, the carrier gas, the flow rate of the carrier gas, and the temperature programming for the isolation process. Hence, the preparation and handling of a column is essential for a successful analysis. The choice for the separation of maternal-fetal metabolites is described below.
Preparation of an efficient packed glass column is necessary for proper partitioning. The column was first filled with chromic acid (with caution while working in a well ventilated hood) and soaked over night. The acid was then drained away and the column was connec ted to tap water and flushed for two hours. It was then sequentially rinsed with 2 liters of high quality deionized water and then 500 ml each of the following nanograde solvents: methanol, acetone, and n- hexane. The column was then placed in a GC oven at 150°C with one end fitted into the injection port while the other end was free. A silylating agent, silyl-8, a gas-liquid-chromatography (GLC) column conditioner (Pierce Chemical Co., Rockford, II.), in 10 ul aliqots was injected onto the column 5 times at ten-minute intervals, with the carrier gas flowing at 5 ml/min. The column was then removed from the oven, both ends were capped and allowed to cool.
After cooling, the column was ready for packing. The packing materials used were 3% 0V-1 on 80/100 Supelcoport, 3% OV-101 on 80/100
Supelcoport or 3% Ov—17 on 80/100 Supelcoport, obtained from Supelco
Inc., Bellefonte, Pa. For the ideal packing of a gas chromatography column (one that gave optimal quality separation), it is important that the material be packed uniformly and tightly into the column. 67
To do this, first a small plug of silanized glass wool was placed into one end of a column. A small funnel was fitted on the other end. The detector side was connected to a vaccum line via a trap. Before turn ing on the vacuum, enough packing material to fill the column should be placed in the funnel at the injector end and then the vacuum was turned on. The suction and gentle tapping of the column enabled the material to pack the column smoothly and continuoussly. If this procedure was not followed, the flow of packing material would stop during the packing process; and once this occurred, it would take a couple of extra hours to pack the rest of the column. This difficulty was probably due to a small extent of moisturization of the packing material as well as the silanized glass wall. As the breaking up of solid support particles would lead to an increase of exposed active sites, which caused trailing of chromatographic peaks, gentle tapping of the column was preferred to the use of vibrator during the packing process. After packing, the suction hose and funnel were removed and the open injector end was plugged with silanized glass wool. The column was ready for conditioning.
Proper conditioning enhances the performance of a column in terms of resolution, prolonged life-span, and low background. One end of the column was fitted onto the injection port while the other end was free to avoid contaminating the interface or the ion sourse of the mass spectrometer. The GC oven temperature was programmed from 30° to 250° C at a rate of 2° /min. The column was then left at 250°C overnight. Helium gas was passing through the column during all stages of conditioning. After conditioning, the column was ready for immedi ate use. Whenever a column was not in use, it ends were tightly capped 68
to exclude moisture. The performance of a column would also improve
if the dryness and purity of the carrier gas were ensured by gas fil
tering cartridges. In the case of a capillary column, in addition to
gas filters, the carrier gas was further purified by immersing part of
the gas carrying copper tubing into liquid nitrogen.
The general procedure for GC/MS analysis of a trimethylsilylated
sample with the Hewlett Packard quadruple instrument was as follows.
By using either the Override or Autotune program supplied by Hewlett
Packard with the GC/MS computer system and using a direct insertion
probe with a known standard, perflourotributylamine (PFTBA), the ins
trument was calibrated and tuned. A representative tuning file con
taining the values of various tuning parameters was shown below.
TUNING FILE
DATE (FRN 1000): FEB 4, 1981 REPELLER(V) =9.44 EM VOLTAGE (V) = 3000 DRAWOUT (V) = 37 AMU GAIN = 134 ION FOCUS (V) = 28 AMU OFFSET = 9 1 ENT LENS (MV/AMU) =140 MASS AXIS GAIN =1.01025 X-RAY (V) = 162 MASS AXIS OFFSET = -1.54218 EMISSION (UA) = 300 IONS: POSITIVE ELECT. ENERGY (EV) = 70 ACTUAL SOURCE TEMP = 200
The system was further checked with a straight chain hydrocarbon
standard mixture containing n-tetradecane, n-hexadecane, n-octadecane,
n-docosane and n-tetracosane as well as trimethylsilyated D-5-mandelic
and n-heptadecanoic acids. The flow rate of the carrier gas (helium)
was adjusted by means of a gas regulator at the supplying gas tank.
When a capillary column was used, the flow rate was further controlled by a valve which adjusted the pressure applied to the GC capillary column. The flow rate was measured by means of a soap bubble meter.
The usual flow rate used for a pack column was 30 ml/min. and that for a capillary column was 4.3 ml/min. Most of the GC/MS anaylses were done in the following conditions. Whenever different conditions were used, they would be specified. The temperatures of the injection port, the jet separator oven (controlled by the auxiliary temperature zone controller in the GC), the GC transfer line probe (controlled by the thermal conductivity detector zone controller in the GC and MS), and the ion source were set at 325, 350, 350 and 200° C, respectively, through the terminal keyboard. For packed columns, on column injection mode was used with the sample size ranged from 1-4 ul. For capillary columns (50m x 0.25mm, glass WCOT, coating CP Jtm Sil 5, Chrompack, The
Netherlands), splitless mode of injection was used with sample size ranging from 0.2 to 3 ul depending on the concentration of samples used. The temperature for GC analyses was generally programmed to start at 70° C, stay at 70° C for 5 min., and then increase to 310°C at 8° per min. The acquisition of mass spectral data was delayed for
3.5-4 mins. so as to vent off the solvent peak(s) to avoid contamination of the ion source. Mass spectral data were acquired with electron impact (El) mode at 70 eV and at an electron multiplier voltage of
3000 V. III. RESULTS
A. Stabilization of Baseline With Open-Spit Interface Helium Preheater
In order to analyze compounds for their elemental compositions, high resolution mass spectrometry was employed. An open split between
the gas chromatograph and mass spectrometer was modified and employed
for the analyses. The open split interface, supplied by the manufac
turer, coupled a Varian Model 3700 gas chromatograph to a Varian Mat
311 A high resolution mass spectrometer. Such an interface was re ported by Henneberg (1975) and Andresen and Ng (1981). The open
split allows compounds, eluting from the end of the GC column, to be
drawn at atmospheric pressure directly into the ion source of the mass
spectrometer through a 0.1 - 0.2 mm I.D. platinum or glass capillary.
This interface has the advantage of requiring no additional vacuum
pumping (in contrast to jet separators). In adition, very intense
GC peaks (such as solvent or unwanted components) can be diverted away
from the open split with a blow out line. This helps to minimize
contamination of the ion source.
However, when attempts were made to utilize the GC-MS system with a capillary column pogrammed from a low (e.g. 60°C) to an elevated
temperature (e.g. 250°C) the total ionization (TI) monitor (and TI plots) drifted significantly below the baseline. Hence, continual
readjustment of the baseline was required during the chromatographic process if the most sensitive TI monitor settings were used. This
70 71 problem was due to the fact that the interface was located inside the
GC oven and was thermally affected by the temperature variations programmed for the oven. During a temperature programmed GC analysis, the amount of purging gas (helium in this case) passing through the interface and entering the mass spectrometer per unit time decreased with increase of the oven temperature. Hence, the baseline drifted down continuously. By keeping the interface at an elevated constant temperature independent of the temperature changes of the GC oven, a steady amount of helium passed into the mass spectrometer and provided a steady baseline.
For this purpose, a small helium preheater was constructed from aluminum blocks which were milled to accommodate stainless steel coils, the open split interface, a heating element and a thermal couple.
The assembly was heated by a heating element with a thermal couple for temperature control. The entirely assembly was thermally insulated from the temperature variation of the GC oven. Hence, the incoming purging gas of helium was heated at constant elevated temperature before entering the open-split interface. The result was that the column eluent, the carrier gas (helium) and the purging gas (helium), the open-split interface, and the inlet line to the mass spectro meter were all at a selected constant, elevated temperature, providing a steady baseline for the TI plots and unperturbed by the temperature programming of the GC oven.
The open-split interface helium preheater was shown in Figures
4-6. The top view and the side view of the preheater were shown in
Figures 4 and 5 respectively. Figure 6 shows an expanded view of the open-split interface arrangement of the capillary from the column, 72
1
Figure 4. Helium preheater and open-split oven: top view, 1) Helium inlet line; 2) aluminum block milled to accept stainless steel tubing coils and open-spit interface; 3) hole to receive GC heating element; 4) hole to receive corres ponding GC thermocouple; 5) glass wool insulation; 6) alignment pins for top aluminum block; 7) T-tube swage-locked to helium gas line and column; 8) threaded hole to receive leg-bolts; 9) aluminum sheet metal enclosure 10) one-quarter to one-sixteenth inch swage-lock with graphite ferrules; 12) mass spectrometer inlet oven. 73
13 TT
Figure 5. Helium preheater and open-split oven: end view. 11)capillary column or cannula from packed column; 12) mass spectrometer inlet oven; 13) wooden block with steel pins to remove top of oven; 14) one-quarter inch bolts; 15) solid 5/8 inch aluminum block with insulation and cover.
Figure 6. Enlargement of open-split interface with alignment jacket: 11) capillary column or cannula from packed column; 16) platinum inlet line to ion source; 17) 21-gauge blow out line; 18) alignment jacket. 74 the platinum inlet line to the ion source, the blow-out line and the alignment sleeve.
The effect of the preheater was illustrated in Figure 7. The upper curve (a) showed the total ionization plot of hydrocarbon standards with a cold open-split interface subjected to the fluctuation of the GC oven with the resultant baseline drift. The lower curve
(b) showed the total ionization plot of hydrocarbon standards when the helium preheater was set at 260°C with a resultant steady baseline.
B. Cleanliness of Instruments and Purity of Reagents
The sucess of GC-MS profiling of biological samples depends heavily on the cleanliness of glassware and the instrument and on the purity of chemicals and solvents. As a preliminary check, an amniotic fluid sample was extracted with chloroform-isopropanol
(3:1) solvent system and was partitioned into neutral, basic and acidic fractions according to the Extraction Flow Chart I (Figure 1).
The acidic fraction was methylated with diazomethane. Then both the acidic and basic fractions were trimethylsilylated with N,0-Bis-
(trimethylsilyl)-trifluoroacetamide (BSTFA) and gas chromatographed with a 16 feet 2 mm I.D. glass column packed with OV-17 on 80/100
Supelcoport as described in Materials and Methods. The TI plots of these three fractions are shown in Figure 8.
Examination of the mass spectral data of the acidic fraction showed that the sample contained a large portion of contaminants in the form of phthalic esters which are generally used as plasticizers in plastics and rubber products. The purity of solvents, the extraction and 75
(b)
Time (min)
Figure 7. Total ionization plots of hydrocarbon standards with: a) cold open-split interface subjected to fluctuations of the GC oven and assoicated baseline drift; b) helium pre heater oven at 260°C without baseline drift. Arrow indicates blow-out of solvent peak for 45 seconds. ACIDS
BASES
NEUTRALS
8835
.1 6345578 91011121314151617l819a«E1222;3242Se7eS2SQ0
Fig. 8 TI. plots of acidic, basic and neutral fractions of amniotic fluid (packed column).
■^j On A.F. ACIDS
DI CHLOROPIETHANE
CHLOROFORfl + ISOPROPANOL
CHLOROFORM + ISOPROPANOL
ACIDS FROM UATER BLANK
12345678 91011121314151617181Sea212;2:E42S2627282a30
Fig. 9 System and solvent check. 78
derivatization procedures were checked for the source of the contamin
ants. Five TI plots are shown in Figure 9. Curve (a) was the tracing
obtained from the methylated and trimethysilylated acidic fraction of
amniotic fluid, showing intense plasticizer peaks at 24-26 minute region.
Curve (b) was obtained by injecting pure dichloromethane alone. The
curve showed that the solvent was acceptable. Curve (c) was obtained by
evaporating 30 ml of chloroform-isopropanol (3:1) by blowing with high
purity N2 gas, dissolving the residue in 5 ul of dichloromethane, and
injecting the solution to the gas chromatograph. Some small ripples were superimposed on the baseline and column bleed. The glass N2 gas
blower which was used to blow dry sample was dismounted and acid washed.
Then 30 ml of chloroform-isopropanol (3:1) was blown dry and subjected to
GC again. As shown in curve (d), contaminants had diminished. Curve
(e) showed the profile of a (non-diazomethane pre-treated) trimethylsily-
lated acidic fraction from extraction of a blank water sample. With the
exception of two peaks from the silylating agent at the 6 minute
region, the system looked clean. From examination of Figure 9, it
became clear that the major phthalic contaminants did not come from
the solvents, buffer or other reagents involved. The next step was
to check the diazomethane methylation process.
Initially the methylation process was achieved by generating
diazomethane in an Erlenmyer flask and bubbling the gas through the
sample to be derivatized via a bent pasteur pipette, one end of
which was joined to the flask with rubber stopper. A diazomethane-methy-
lated and BSTFA-trimethylsilylated acid fraction of blank water extract
showed heavy contamination with phthalic ester. This was obviously due
to the carrying over of plasticizers by the ether vapor from the rubber 79 stopper as the diazomethane gas bubbled into the sample tube. The diazomethane generating apparatus was redesigned to eliminate using any rubber stopper or rubber tubing. The final methylation set up was shown in Figure 3. The major contamination by phthalic esters during methylation was eliminated.
C. Capillary GC-MS Profiles of Acidic Fraction of Amniotic Fluid
Thirty-four biological samples were extracted, and nineteen of them were partitioned into various fractions according to the Extraction
Flow Chart 2 (Figure 2). These samples included amniotic fluid (15), ovarian cyst fluid (1), serum (2), and meconium (1). These samples were analysed with a capillary column. Fourteen of the capillary GC-MS total ion (TI) plots plots are discussed in detail below. Other TI plots are documented in the Appendix. A list of the sample and patient information is shown in Table 7. Unless otherwise indicated, the samples used are amniotic fluid.
Sample 1 was obtained from a patient age 37. Amniocentesis was performed at 15.5 weeks of pregnancy because of maternal age. Medica tion included vitamins plus iron. The fluid was yellow. Centrifugation produced very little precipitate. The GC-MS profile of the acidic fraction (El) is shown in Figure 10. Most of the major identified peaks are labelled in the figure. Minor identified peaks are numbered.
A listing of the identified components and the corresponding GC peak number is given in Table 8. Twenty-two components of this sample were identified. Phenyl-d5-mandelic acid and heptadecanoic acid were used as internal standards (each at a concentration of 1.5 ug/ml).
The four major components were lactic acid, 2-keto-isocaproic acid, urea 80
Table 7. List of capillary GC-MS analysed samples and patient informations.
Patient Patient Preg. wk. Comments No. name
1 JNS 15.5 for maternal age, vitamins, Fe
2 AIB 16 for maternal age
3 NBG 38.5 no medication, repeat Cesarean section
4 DLS 15 open tubular neural defects
5 MLD 30 severe Rh, phenytoin
6 BB 37 Rh sensitized, 0D450= .01
7 CM 16-18 nicotine 0.5 pk/day tylenol, vitamin
KRG 38 Delalutin q wk, tapped for L/S
9 DL Hydrops
10 MDH 35 diabetic
11 JM 31 Down syndrome & GI block
12 MZ 22 drug addict
13 DLS NA blood sample
14 MAR NA Ovarian cyst fluid 81
Table 8. List of identified compounds with the corresponding GC peak numbers.
Compound £ Compound £ lactic acid-di-TMS 2 p-OH-phenylacetic-di-TMS 42 2-OH-butyric-di-TMS 7 lauric-TMS 44 3-0H-butyric-di-TMS 8 homovanillic-di-TMS 46 3-0H-isobutyric-di-TMS 9 hippuric-di-TMS 48 2-0H-2-methyl-butyric-di-TMS 10 hippuric-TMS 49
2-0H-isovaleric-di-TMS 12 myristic-TMS 51 acetoacetic-di-TMS 13 vanillylpropionic-di-TMS 52 benzoic-TMS 14 indole-3-acetic-di-TMS 54 urea-di-TMS 15 pentadecanoic-TMS 56 2-keto-isocaproic-di-TMS 17 palmitoleic-TMS 58
2-0H-caproic-di-TMS 18 palmitic-TMS 60 2-keto-valeric-di-TMS 19 indole-propionic-di-TMS 62 2-keto-isovaleric-di-TMS 21 heptadecanoic-TMS (IS) 64 octanoic-TMS * 22 linoleic-TMS 68 phenylacetic-TMS 24 oleic-TMS 69 phosphoric-tri-TMS 25 stearic-TMS 70 2-keto-3-methyl-valeric-di-TMS 27 dioctyladipate 73 2-ke to-caproi c-d i~TMS 28 dioctylphthalate 74 nonanoic-TMS 29 squalene 75 decanoic-TMS 30 cholesterol-TMS 80 phenyl-d5-mandelic-di-TMS (IS) 32 salicylic-di-TMS t 34 phenyllactic-di-TMS 38 p-OH-benzoic-di-TMS 39 4-acetylamino-phenol-di-TMS 40
* found only in serum t found only in ovarian cyst fluid TI
Fig. 10 TI plot of trimethylsilylated CH2CI2 extract of acidic components of AF(E1): (15.5 weeks, tapped for maternal age). and cholesterol. Although cholesterol is generally considered to be a neutral molecule, it nevertheless appeared in all acidic frac tions, because the solvent partition method depends on the concentration of the component and its partition coefficients between the solvent systems. If a given compound is present is a large quantity relative to the amount of solvents used for the partitioning process, the compound is seldon extracted exclusively into only one of the solvent phases and none in the other. The organic acids consisted mainly of straight chain fatty acids (nonanoic, decanoic, lauric, palmitic, linoleic, oleic, stearic), keto-acids (2-keto-isocaproic, 2-keto-caproic), short chain hydroxyfatty acids (lactic, 2-OH-butyric, 3-OH-isobutyric,
2-0H-2-methylbutyric), aromatic acids (phenyllactic, hippuric). Urea and phosphoric acid were also identified in the profile.
Sample 2 was obtained by amniocentesis at the 16th week of ges tation because of maternal age. The specimen was yellow. Centrifu gation produced little precipitate. Cell counts and protein in amnio tic fluid are usually lower in early gestation. The GC-MS profile of the acidic fraction (El) is shown in Figure 11. The four major com ponents are lactic acid, 2-keto-isocaproic, cholesterol, and urea.
The identified organic compounds included long chain fatty acids (non anoic, decanoic, lauric, myristic, palmitoleic, palomitic, linoleic, oleic, stearic), keto-acids (2- keto-isocaproic, 2-ketoisovaleric,
2-keto-caproic), hydroxy-acids (lactic, 2-OH-butyric, 3-OH-isobutyric,
2-OH-methyl-butyric), aromatic acids (phenyllactic, hippuric, homovani- llic), phosphoric acids, and urea. In addition, dioctyladipate and dioctylphthalate were identified. The latter two compounds are plasti- cizers. On comparison of Figure 10, 11, and the later samples, it is 68
TI r~T- r r-r
Fig. 11 TI plot of trimethylsilylated CH2C12 extract of acidic components of AF(E1): 16 weeks, tapped for maternal age • not clear whether these two compounds were actually present inside the amniotic cavity through maternal exposure to these compounds, or were artifacts introduced by sample handling procedures or a combination of
both. From previous experience of extraction and derivatization of blank water sample, the amount of phthalate detected was insignificant, and dioctyladipate was not detected.
Figures 10 and 11 represent two samples of amniotic fluid obtained from early pregnancy not associated with particular medication or medical problem. Although two samples were not sufficient to define any physiological condition(s), it was clear by comparing the two profiles that they were of a similar pattern. First of all, lactic acid was the most intense peak in both samples. The size of the lactic acid peak relative to the internal standard, phenyl-d5-mandelic acid, was the same in each profile. The retention times of the various peaks relative to the internal standard were similar in both samples. The partially resolved doublets of the peaks of 3-OH-isobutyric and 2-0H-
2-methyl-butyric eluting at 9.1-9.3 minutes; doublets of 2-keto-3- methyl-valeric and 2-keto-caproic eluting at 12.3-12.5 minutes, the pattern at 16.5-17.5 minutes with two unknown peaks between phenyl lactic and lauric acids; the cluster of peaks for linoleic, oleic and
stearic acids; the relative intensities of urea, palmitic, cholesterol, all substantiated the appearance of a similar profile. It can also be pointed out that for these two amniotic fluid samples obtained at 15.5 and 16 weeks of pregnancy, the amount of hippuric acid was relatively
small.
Sample 3 was an amniotic fluid specimen obtained at 38.5 weeks of pregnancy by Cesarean Section. The patient had no medical problem TI
Fig. 12 TI plot of trimethylsilylated CH2C12 extract of acidic components of AF(E1): (38.5 weeks no medication, repeat Cesarian section). 87 and received no medication. For this sample, phenyl-d^-mandelic acid was not used as internal standard. Heptadecanoic acid at 1 ug/ml
(peak 64) was the only internal standard. The acidic fraction profile was shown in Figure 12. The T1 plot showed that lactic acid was most
abundant. Other major constituents were homovanillic, hippuric and
cholesterol. The minor components included 2-OH-butyric, 3-0H-
butyric, 2-0H-2-methyl-butyric, benzoic, phosphoric, p-OH-phenyl-
acetic, myristic, pentadecanoic, palmitoleic, palmitic, heptadecanoic
(I.S.), linoleic, oleic, and stearic. The interesting feature of this
sample as cmpared to Samples 1 and 2, was that it contained a much
larger quantity of hippuric acid, and homovanillic acid, with a re
duction of 2-keto-isocaproic, and a replacement with 3-OH-butyric for
3-OH-isobutyric. An increase of hippuric acid might be a reflection
of the maturity of the fetal kidney function at late stage of pregnancy.
Sample 4 was obtained by amniocentesis from a patient at the 15th
week of pregnancy with a fetus suffering open tubular neural defect.
The acidic fraction profile of this amniotic fluid sample was shown in
Figure 13. The four major identified components were lactic acid, urea,
2-keto-isocaproic, cholesterol and palmitic. The level of hippuric acid
was similar to those of samples 1 and 2 (at 15.5 and 16 weeks) and
far less than that of sample 3 (at 38.5 weeks). Compared to samples
1-3, it had a higher level of urea. Interestingly, it contained indole-
3-acetic acid, whose peak size was well above the baseline. The pre
sence of the indole-3-acetic, together with cluster of peaks around the
indole-3-acetic, and the tentative identification of squalene eluting
at 29 minutes seemed to characterize this acidic fraction.
Sample 5 was obtained by amniocetesis from a 30 week old pregnancy ) c0 O rHV •O <9 c 3 •H D* I ^<0 I I I ^
TI , 46 56 f 64 69 --- ' ------i—t— i— r—t— i— r— rn— i i— i— i— i— i— i— i—t— . "1 1 1 1 1 1 13-1313 15 16 17 18 l!j gB SI 2S S3 ci4 £5 S6 g7 28 £9 30 31 73d. 33 33 35 56 37 38
Fig. 13 TI plot of trimethylsilylated CH2CI2 extract of acidic components of AF(E1): (15 week, open tubular neural defects). |V 28
TI i— r~ — i— r i— r
Fig. 14 TI plot of trimethylsilylated CH2CI2 extract of acidic components of AF(El): (30 weeks, severe Rh-immunized, Phenytoin 100 mg bid). 90 complicated by severe Rh-immunization. The acidic fraction profile was shown in Figure 14. The major peaks were lactic, hippuric, cholesterol, urea and 3-0H—butyric. The interesting features of this profile were the relatively large quantity of hippuric acid and the presence of a sizeable peak of indole-3-aacetic acid. The presence of urea, 2-keto-isocaproic, the doublet of 2-keto-3-methyl-valeric and 2-keto-caproic, the presence of nonanoic, deanic, palmitoleic, palmitic, linoleic, oleic, and stearic rendered it the general pattern of an amniotic fluid acid profile. The large hippuric peak was con sistent with late gestation period. The presence of a large amount of indole-3-acetic acid might have some bearing on the stress to the fetus.
Sample 6 was obtained by amniocentesis from a 37 week pregnancy complicated by Rh sensitization. The specimen appeared light yellow. The acidic fraction profile was shown in Figure 15.
This TI plot showed an unusually high quantity of dioctyladipate
(peak 73) and dioctylphthalate (peak 74). The other major peaks were lactic acid, hippuric, palmitic, and cholesterol. The hump eluting between 27-30 minutes were mainly plasticizers. Features of biological interest were the high level of hippuric acid at 37 weeks of gestation, the presence of a large level of indole-3- acetic adjacent to an intense unknown peak which eluted at 20 minutes.
Sample 7 was obtained by amniocentesis from a pregnancy of a gestational age estimated to be 16-18 weeks. The patient was known to smoke half a pack of cigarettes per day, and was taking
Tylenol and vitamins. The acidic fraction profile was shown in TI
Fig. 15 TI plot of trimethylsilylated CH2CI2 extract of acidic components of AF(El): (37 weeks, Rh sensitized, OD450 = 0.01). cholesterol
Fig. 16 TI plot of trimethylsilylated CH2CI2 extract of acidic components of AF(El): (16-18 weeks, nicotine 0.5 pack per day, Tylenol, vitamins). Figure 16. The major peaks were lactic, 2-keto-isocaproic,
cholesterol, phosphoric. The presence of a large peak of indole-
3—acetic acid was most unique to this profile. The other unusual
features of this fraction were the relatively higher levels of
phosphoric acid, 2-keto-isocaproic acid, and the presence of
2-keto-valeric acid. Although no heptadecanoic acid was used in this
sample as internal standard, this acid was detected (peak 64).
The amount of hippuric acid was consistent with the early gestation
period.
Sample 8 was obtained by amniocentesis at the 38th week of
pregnancy. The patient was given Delalutin once a week. The sample was white, containing a lot of floating particles of cellular debris.
The GC-MS profile of the acidic fraction was showed in Figure 17.
This fraction contained a high level of dioctyadipate and some
phthalates in the region eluting between 25-32 minutes. The major
components were lactic, urea, homovanillic, hippuric, palmitic and a
large amount of unidentified steroid eluting at 29.5-29.7 minutes.
Examination of the mass spectral data revealed some unusual phenomena.
Urea appeared to spread through the entire region between 10.3-11.5
minutes. Some phosphate-trimethylsilylated ions was overlapped within
the urea peak at 11.3 minute region. Possible explanation was that a
large quantity of urea was present in the sample. This might overload
the capillary column and urea behaved as a stationary phase to the
other components eluting close to that region. Furthermore, most of
the hippuric acid in a given sample usually appeared as di-trimethyls-
ilylated and only a small portion was mono-trimetylsilylated. In
this particular sample, the hippuric acid appeared as a doublet of i. 7 TIplotoftrimethylsilylated CH 17Fig.
2-OH-burvric (38 weeks,Delalutinonce a tappedwk, for L/S). 1 T -- 1 2 CI 2 extractof acidic components ofAF(El): ■* r~ equal intensity. Nevertheless, hippuric was a major component in this profile, consistent with its gestational age. The relative intensity of homovanillic acid also deserved noting.
Sample 9 was obtained from a pregnancy complicated with hydrops.
The GC-MS profile of the acidic fraction was shown in Figure 18. The most intense peak was lactic acid. Two other intense peaks were 3-0H- isobutyric and an unidentified peak eluting at 23.9 minutes. Indole-
3-acetic acid appeared as a very small peak above the baseline. There were moderate peaks for acetoacetic and homovanillic acid. The gesta- tional age of this sample was not known. However, judging from the amount of hippuric acid present, it could be estimated to be within the range of 15-20 weeks old pregnancy, unless hydrops or other conditions caused a decrease of hippuric acid in amniotic fluid.
Sample 10 was obtained by amniocentesis from a pregnancy of 35 weeks and complicated with diabetes. The acidic fraction profile is shown in Figure 19. The most abundant component was lactic acid.
The next intense peak was hippuric acid. Most of the other peaks appeared small. Urea was eluting at an unusual place relative to phosphoric acid. The presence of a large peak of hippuric acid was consistent with the gestational age, however, it occurred almost entirely as mono-trimethylsilylated instead of the usually observed di-trimethylsilylated. The cholesterol peak was unusually small com pared with other samples. 2-0H-butyric, 3-0H-isobutyric, acetoacetic and homovanillic were significant relative to other long chain fatty acids. There was a small peak of indole-3acetic acid, barely visible above the background. > -H
58. Aa-AAv. 38 .69. i— r
Fig. 18 TI plot of trimethylsilylated CH2CI2 extract of acidic components of AF(E1): (hydrops).
VO On 28
25 TI — r i— r t -r -=-i— ,
Fig. 19 TI plot of trimethylsilylated CH2C12 extract of acidic components of AF(El): (35 weeks, diabetic). 98 Sample 11 was obtained by amniocentesis from a pregnancy of 31 weeks and complicated with Down syndrome and gastrointestinal block.
The GC-MS profile of the acidic fraction is shown in Figure 20.
Similarly, the most intense peak is lactic acid. The other major components were hippuric acid, cholesterol, urea, and palmitic acid.
A major compound, 4-acetyl-aminophenol, which was not observed pre viously in the other samples was presnet in this sample. Whether this drug had any connection with the Down syndrome and gastrointestinal
block or its presence was a mere coincidence is not known. It would
seem premature to imagine a single sample could implicate any special physiological conditions. Yet, the result might suggest some
possible association between the presence of this compound and the
the abnormality. With the exception of the intense peak of 4-acetyl- aminophenol, this profile was very similar to other profiles of late gestation with the presence of clusters and doublets of short chain hydroxy- or keto- fatty acids in the 6.5-11 minutes region, and the
scattering of lauric, myristic, palmitolenic, palmitic, linoleic,
oleic and stearic over the 16-24 minutes region. The homovanillic
peak appeared higher than the others.
Sample 12 was obtained by amniocentesis from a pregnancy of 22 weeks. The patient was addicted to pentazocine, and suffered
from malnutrition and anemia. The GC-MS profile of the acid fraction
is shown in Figure 21. Similar to the profiles shown earlier, the
relative intensity of lactic, cholesterol, 2-0H- butyric, 2-keto-
isocaproic, and the scattering of the long chain saturated and unsatur" ated fatty acids were also observed. However, hippuric acid was not
observed in this sample. The unusual feature of this profile was the 25 68 TI
Fig. 20 TI plot of trimethylsilylated CH2CI2 extract of acidic components of AF(El): (Down syndrome and GI block). 1 i— i— i— r
Fig. 21 TI plot of trimethylsilylated CH2C12 extract of acid components of AF(El): (pentazocine, malnutrition, anemia). 101 very intense peaks of 3-0H-butyric and acetoacetic, and an unidentified
peak at 8.2 minute. The presence of the large 3-0H-butyric and aceto
acetic peaks might be a reflection of the nutritional status.
D. Capillary GC-MS Profile of Acidic Fraction of Serum
Sample 13 (see Table 8) was a serum sample obtained from a patient who had a pregnancy complicated with open tubular neural defect (same
patient as Sample 4). The GC-MS profile of the acid fraction was
shown in Figure 22. This profile, as was expected, differed from all
the other profiles of amniotic fluid. This specimen contained rela
tively less lactic acid and cholesterol. There were some unidentified
major peaks eluting before lactic acids. Octanoic acid, which was not
detected previously in other samples of amniotic fluid, was found in
this serum. The pairs of doublets (3-0H-isobutyric and 2-OH-2-methyl-
butyric; 2-keto-3-methyl-valeric and 2-ketocaproic) were present, but
of a less intensity than those observed in the amniotic fluid. Linoleic
and oleic acid were more abundant than stearic acid in the serum sample, while the reverse was true with amniotic fluid samples. In general,
the fatty acids seemed more intense than the hydroxy- and keto- acids. U OJ •H T3 o c tiC <30 I u 2 I
VI VJ 64 ?5 “19 LLxi UAJ TI ~ i i i t i i r I i i r t r i i i i t i i i i r [—r—r —i—i— i—i— i—i—r a____2 _'d__'j 10 11 15 13 14 15 16 1? 13 15 dO 51 £3 3 3 £4 £5 5S £? 58 59 30 31 38 33 34 35 36 37 38
Fig 22 Plot °f trimethylsilylated CH2CI2 extract of acid components(E1) of serum from pregnancy with open tubular neural defects. 103 E. Capillary GC-MS Profile of Acidic Fraction of Ovarian Cyst Fluid
Sample 14 (see Table 8) was an ovarian cyst fluid obtained from a
31 year old patient who was diagnosed as non-fertile and had premature menopause. The sample was a yellowish-green viscous fluid which
fluoresced in long wavelength ultraviolet light. After centrifugation
at 10,000 xg for 20 minutes, a band of material appeared at the top of
the tube gave a bright-bluish fluorescence under short wavelength
ultraviolet light. The sample was so viscous that it had to be diluted with an equal volume of pure water before the extraction according to
the procedure listed in Extraction Flow Chart 2. The GC-MS profile of
the acidic fraction was shown in Figure 23. This Profile was quite
different from those of amniotic fluid or serum. The components were
similar, but their relative concentrations were very much different.
Cholesterol appeared most abundant. Long chain saturated and un
saturated fatty a ■'ds dominant the profile. There was an unusual
large peak of salicylic acid in this profile. It was not clear how
salicylic acid get transported into an ovarian cyst. The phenyllactic
acid peak was relatively large as compared to those of Samples 1-13.
The heptadecanoic acid peak was unexpectedly large. This observation
might suggest that the ovarian cyst fluid contained a considerable
amount of endogenous heptadecanoic acid, since a small quantity of
this compound (1.5 ug/ml) was added to the diluted sample (1:1) of
the ovarian cyst fluid as an internal standard (Phenyl-d5“mandelic
acid (peak 32) was also added at 1.5 ug/ml as an internal standard).
This is further supported by the fact that when no heptadecanoic
acid was added as internal standard in Sample 7 (Fig. 16), this com
pound was also detected. This strongly suggested that Sample 7 also 28 30
TI
Fig.23 TI plot of trimethylsilylated CH2CI2 extract of acidic components(El) of ovarian cyst fluid. 105 contained endogenous heptadecanoic acid. The cluster of hydroxy- and keto- acids were at the initial portion of the TI plot, though their relative intensities were different from those observed in amniotic fluids.
F . Comparison of Organic Acids Identified in this Study with those
Identified by GC-MS Profiling of Amniotic Fluid in the Literature
A list of organic acids identified in this study and those identified by GC-MS profiling of amniotic fluid in the literature is given in Table 9.
The "+" sign indicates a particular compound was identified by the corres ponding group of investigators. A comparison between the columns of Table 9 shows that dicarboxylic acids (e.g. oxalic, succinic, maleic, fumaric, malic, 2-ketoglutaric), tricarboxylic acids (e.g. citric, isocitric), and hydroxylated dicarboxylic acids (e.g. maleic, 3-hydroxy-3-methyl- glutaric, 2-hydroxy-glutaric) were not identified in the present study.
However, a number of hydroxy- and keto- monocarboxylic acids (e.g. 2-0H-
2-methylbutyric, 2-OH-isovaleric, acetoacetic, 2-OH-caproic, 2-ketovaleric,
2-ketoisovaleric), fatty acids (e.g. nonanoic, decanoic, palmitoleic) and aromatic carboyxlic acids (benzoic, phenylacetic, phenyllactic, indole-3- acetic, indole-propionic, vanillylpropionic) have been identified under the current investigation but not in the GC-MS profiles reported in the literature (Hagenfeldt and Hagenfeldt, 1972; Williams et al, 1979; Nicholls et al, 1978; Shinka et al, 1979).
There can be several reasons for the differences.
(1) Dichloromethane was used as the organic extraction solvent for the 106 the present study, while ethyl acetae, diethyl ether/ethyl acetate or silica gel were used by other investigators. Different sovents tend to extract different classes of compounds both quantitatively and qualita tively. (2) Protein was not removed by deproteinization procedures before extraction in the present study. It is possible some dichoro- methane-extractable organic acids are loosely bound to protein in the amniotic fluid. (3) A 50m x 0.25mm WCOT capillary column was used in this study. The much higher resolving power of capillary column compared with that of packed columns facilitated the identification of components of amniotic fluid.
Nicholls et al reported profiling 40 samples of amniotic fluid which fell into two main gestational groups, 15-20 weeks and 30 weeks-term.
They have shown a typical metabolic profile of organic acids from amniotic fluid. With my present method of investigation, I have found an obvious change in the profiles between the gestation age of 15-20 weeks and 30 weeks-term. The former group has less hippuric acid content relative to other components in the profile, while the latter group has significantly higher hippuric acid content. This may reflect the maturity of fetal kidney function.
No GC-MS profiling of amniotic fluid organic acids from diseased state pregnancies has appeared in the literature. The profiles obtained in this study from pregnancies complicated with open tubular neural de fects, Rh sensitization, diabetes mellitus, drug addiction and malnutri tion, and smoking are intriguing and deserve further investigation. 107
Table 9. Organic acids identified by Williams et al 1979, Nicholls et al 1978, Shinka et al 1979 and Ng et al 1981.
Organic Acids Williams Nicholls et al Shinka Ng et al Wilkinson et al et al et al glycolic + + lactic + + + + pyruvic + + 2-hydroxybutyric + + + + 3-hydroxybutyric + + + + 3-hydroxyisobutyric + + 3-hydroxyisovaleric + 3-hydroxypropionic + cis-glyceraldehyde + trans-glyceraldehyde + oxalic + urea + + + 2-ketocaproic + + 2-ketoisocaproic + + dihydroxyacetone + phosphoric + + + + succinic + + + tartronic + maleic + glyceric + + 2-methy1-3-hydroxy- + butyric fumaric + 2-deoxytetronic + lactic dimer + malic + + 2-hydroxybenzoic + + 2-hydroxyphenylacetic + 2-hydroxyglutaric + + 3-hyd r oxy-3-methy1- + glutaric 4-hydroxybenzoic + + 3-hydroxyphenylacetic + 4-hydroxyphenylacetic + gentisic + pyroglutamic + lauric + + + alpha-glycerophosphate + beta-glycerophosphate + 2-ketoglutaric + + homovanillic + + 108 Table 9. Organic acids identified (continued)
Organic Acids Williams Nicholls et al Shinka Ng et al Wilkinson et al et al et al citric + + myristic + + + vanillic + isocitric + fucono-1,4-lactone + 4-hydroxyphenyllactic + + glucuronolactone + palmitic + + + hippuric + + heptadecanoic + + + dodecanedioic + uric + oleic + + + linoleic + + stearic + + + + arachidonic + 2-OH-2-methylbutyric + 2-OH-isovaleric + acetoacetic + benzoic + 2-OH caproic + 2-keto-valeric + 2-keto-isovaleric + octanoic + phenylacetic + 2-keto-3-methyl-valeric + + nonanoic + decanoic + phenyllactic + 4-acetylaminophenol + indole-3-acetic + palmitoleic + indole-propionic + vanillylpropionic IV. CONCLUSIONS
A. A unique method of extracting amniotic fluid and other biological samples was developed. Dichloromethane was used as the extraction solvent. It eliminated the problem of possible esteri- fication or tranesterification with fatty acids or esters. However, the above processes occurred when chloroform-isopropanol or ethyl- acetate were used for extraction. The current method did not involve precipitation of protein with reagents like cold ethanol, perchloric acid, or sulfosalicylic acids. This had the advantage of including in the profile metabolites which were loosely associated with protein and extractable from the sample. As the solubility of dichloromethane in water was lower than that of other solvents like ethyl acetate, com pounds with low water solubility would be removed from the aqueous sample. This method produced a lower background in the gas chroma togram or total ion plots as compared with those obtained from ex traction with ETOAC. The extraction procedure, shown in the
Extraction Flow Chart 2 (Fig. 2), has the advantage of simplifying the gas chromatogram by separating the relatively less polar components with dicholoromethane extraction and leaving the remaining compounds to be extracted with ethyl acetate. However, if organic acids were the only interest in the profiling of amniotic fluid, there are many other more quantitative methods for removing organic acids from the sample, e.g. DEAE anion exchange chromatography, adsorption and desorption from silica gel. The latter two methods do not allow profiling of the neutral and basic components of the same sample.
109 110 The extraction procedure outlined in Figure 2 included acidic hydro lysis process after initial removal of free neutral, basic and acidic components. The present method allows profiling of conjugated com ponents in the same sample as well. Hence, the extraction procedure will provide fractions all of which will contribute to the chemical picture of a given amniotic fluid sample. Such a composite picture will contribute to a more complete understanding of the environment and condition of the developing fetus.
B. In the process of developing methodologies for GC-MS profiling of amniotic fluid, two simple designs were invented. One was applied to the methylation of micro sample of organic acids with diazomethane.
The other concerned the stabilization of the baseline of the total ionization plot during a temperature programmed capillary GC-MS analysis.
Diazomethane is known to be an explosive gas. The generation of diazomethane should be carried out in smooth glass system. Hence, ground glass joints, any flaw or scratch in glasswares should be avoided. By applying a simple glass-blowing technique which took less than five minutes, a delivery tube was made. This newly designed glass tube together with a 50 ml erlemyer flask, and a half inch of teflon tubing became a very useful setup for generating diazomethane and derivatizing organic acids (see Figure 3 ).
The problem of baseline drifting associated with the use of the open-split interface supplied by Varian to couple the Varian
Model 3700 gas chromatograph to the Varian MAT 311A high resolution mass spectrometer had been discussed in the previous section. As a solution to the problem, a helium preheater was designed and cons tructed. The design was shown in Figures 4-6 and a patent is pending. Ill
C. The GC-MS profiling of organic acids of amniotic fluid with glass packed column was not very satisfactory. This was due to the complexity of biological sample and the limited resolution inherent with a packed column. After all details of quality control concerning solvents, reagents, extraction, derivatizating, sample handling, instru ment repair maintenance and operation had been worked out with packed column, capillary column was used for profiling of organic acids from amniotic fluid. Nineteen biological samples were extracted according to procedure outlined in the Extraction Flow Chart 2 (Figure 2).
At present, fourteen GC-MS profiles of organic acid fractions (El) have been examined in detail, and the following conclusions can be made.
Capillary GC-MS profiling of amniotic fluid provides a multi variate composite picture of the chemical environment in which the development fetus is suspended. Other methodologies cannot provide such vast spectrum of chemical information with such high specificity and precision.
On careful examination of the profiles, the corresponding mass spectral data, it is evident that a pattern emerged to characterize the acid fraction of an amniotic fluid. There are several markers to outline the profile. First, there is lactic acid at the beginning of the profile, which in most instances is the most intense peak.
This is followed by a cluster of short chain hydroxy- and keto- fatty acids. 2-OH-butyric acid is normally a small isolated peak.
Then there are two doublets, the first consists of 3-OH-isobutyric and 2-OH-2-methyl-butyric, the second doublet consists of 2-keto-
3-methyl-valeric and 2-keto-caproic. In between these two doublets urea, 2-keto-isocaproic, and phosphoric acid are found. The selection of phenyl-d5_mandelic acid as an internal standard is most appropiate, since it elutes in a region where no major peaks occur. It provides a good reference point for calculating the relative retention indices of the eluting components and for locating minor fatty acid components such as nonanoic and decanoric acids. In the center of each profile, in the region around 17-18 minutes, there is normally fairly large peak which has not be identified. It is actually composed of more than one compound and gives a somewhat flatten peak. This serves as a landmark in this region around which phenyllactic and lauric acids are often detected. The next peak is hippuric acid which occurs in di- and mono- trimethylsilylated form. From all the samples examined so far, the quantity of hippuric acid reflects the duration of the pregnancy. Samples from early gestation have less hippuric acid (as illustrated in Figures 10, 11, 13, 16, 21), while samples from late gestational period have a higher content of hippuric acid (Figures 12,
14, 15, 17,19). The other characteristics of the profile are the location of the various long chain fatty acids, such as palmitoleic and palmitic, and the grouping of linoleic, oleic and stearic. Finally, the cholesterol peak is detected at the end of the profile and other steroids are also detected around this end region.
One particular endogenous metabolite deserves special notice.
Indole-3-acetic acid was either not detected or existed in minute quantity in most of the profile. However, three samples contained a significant amount of indole-3-acetic acid. The amniotic fluid sample which contained the largest quantity of indole-3-acetic acid was obtained from a patient who smoked half a pack of cigarettes a day and received Tylenol and vitamins. The other two samples, obtained from pregnancy complicated with Rh sensitization, also con tained a comparatively much larger quantity of indole-3-acetic acid than that found in other samples. However, it is not clear whether the occurrence of a high level of indole-3-acetic acid in amniotic fluid has any biochemical bearing with Rh sensitization or with smoking in pregnancy. Definitely, the coincidence is intriguing.
The GC-MS profile of Sample 12 is also interesting. It has the common markers of amniotic fluid acidic fraction, namely, the intense lactic and cholesterol peaks, the grouping of hydroxy- and keto- fatty acids after lactic acid, and the distribution of other fatty acids throughout the profile. It has a large quantity of 3-0H- butyric (not observed in other samples), and a relatively large amount of acetoacetic. The peak height for the other fatty acids are rather small. However, hippuric acid is not observed. This profile is consistent with the clinical assessment of malnutrition. Possibly, the absence of hippuric acid might reflect poor development of the fetal kidneys or their functions.
GC-MS profiling is definitely a powerful method for assessment and understanding of the fetal environment. With the advance of medical science, more and more individuals who might not have lived to their adult years in decades ago because of genetic diseases, do presently live to produce offsprings. An increasing percentage of the population will have genetic disorder of one form or another. Besides, it is estimated that 5000 new chemicals are introduced every year, and only a very small portion of these new chemicals have been tested for carcinogenicity, mutagenicity and teratogenicity. The human genetic pool presumably is under constant assault. It is obvious that the necessity for prenatal diagnosis is urgent, not to search for deformity and destroy, but to detect the problem and device clinical measures to prevent the formation of defects or deformities. GC-MS profiling of amniotic fluid would be a useful instrument towards that goal. V. BIBLIOGRAPHY
Abramovich, D.R., and Page, K.R. (1972): Pathways of water exchange in the fetoplacental unit at mid pregnancy. J. Obstet. Gynaec, Brit, Cwlth. 79, 1099
Abramovich, D.R. (1970): Fetal factors influencing the volume and composition of liquor amnii. J. Obstet. Gynaec. Brit. Cwlth. 77, 865
Abramovich, D.R. (1968): The volume of amniotic fluid in early preg nancy. J. Obstet. Gynaec. Brit. Cwlth. 75, 728
Ampola, M.G., Mahoney, M.J., Nakamura, E., Tanaka, K.(1975): Prenatal therapy of a patient with vitamin-Bi2-resPonsive methylmalonic acidemia. 293(7), 313-7
Andresen, B.D., Ng, K.J., Wu, J., Bianchine, J.R. (1981): Helium pre heater for the open split interface. Biomed. Mass Spec, (in print).
Anteby, S. 0., Zukerman, H., Gedelia, I. and Weistreich, V. (1973): Citric acid in amniotic fluid. 80, 27-8
Anton, A.H. and Sayre, D.F. (1971): The distribution of 5-hydroxyindole- acetic acid in various animals and its fluorometric analysis in diverse biological material. J. Pharm. Expert. Therap. 179(2), 207-17
Bain, A.D., and Scott, J.S. (1960): Renal agenesis and severe urinary tract dysplasia. Brit. Med. J., 1, 841
Balint, J.A., Kyriakides, E.C., Gunawardhane, G.D.V., Risenberg, H. (1978): Surfactant lecthin fatty acid composition and its relationship to the infantile respiratory distress syndrome. Pediat. Res. 12, 715-9
Belisle, S., Tulchinsky, D. (1980): Amniotic fluid hormones. In: Mater nal-fetal endocrinology. Tulchinsky, D., and Ryan, K.J. (Eds.) W.B. Saunders Co. Philadelphia Pa.
Bichler, A., Daxenbichler, G., Geir, W. (1977): Determination of amnio tic fluid palmitic acid concentration for the estimation of fetal lung maturity. Clin. Chim. Acta, 74, 133-138
Biezenski, J.J. (1973): Origin of amniotic fluid lipids. III. Fatty acids. Proc. Soc. Expt. Biol. Med., 142, 1326-8
115 116 Brock, D.J.H.; Bolton, A.E.; Monaghan: Prenatal diagnosis of anence- phaly through maternal serum alphafetoprotein measurement. Lancet 2 (1973) 923
Brock, D.J.H. and Sutcliffe, R.G. 1972: Alpha-fetoprotein in the antenatal diagnosis of anencephaly and spina bifida. Lancet ii, 197
Brock, D.J.H., Hayward, C., Seppala, M. (1980): Distinguishing neural tube defects and congenital nephrosis by amniotic fluid assay. Lancet, 1(8171), 1773
Brock, D.J.H. (1979): Foeto-specific proteins in prenatal diagnosis. Biochem. Soc. Trans. 7(6), 1179-95
Brzezinski, A., Koren, Z., Pfeifer, Y., and Sulman, F.G. (1962): The metabolism of serotonin in amniotic fluid. J. Obstet. Gynec. Brit. Cwlth. 69, 661-4
Burlingame, A.L., Kimble, B.J., Derrick, P.J. (1976): Mass spectrometry. Anal. Chem. 48(5), 368R
Burton, B.K., Gerbie, A.B., Nadler, H.L. (1978): Enzymes in the amniotic fluid. In: Amniotic fluid— Research and clinical application. 2nd rev. ed. Fairweather, D.V.I.; Eskes, T.K.A.B. (Eds.), Excerpta Medica p. 225
Campbell, S., Wladimiroff, J. W. and Dewhurst, C.J. (1973): The ante natal measurement of fetal urine production J. Obstet. Gynaec. Brit. Cwlth., 80, 680
Carr, D.H. (1969): Chromosomal errors and Development. Amer. J. Obstet. Gynec., 104, 327
Chan, W.Y., Seale, T.W., Shukla, J.B., Rennert, O.M. (1979): Polyamine conjugates and total polyamine concentrations in human amniotic fluid. Clin. Chim. Acta 91, 233-241
Charles, D., Jacoby, H.E. and Burgess, F. (1965): Amniotic fluid volumes in the second half of pregnancy. Amer. J. Obstet. Gynec., 93, 1042
Clements, J.A., Brown, E.S., Johnson, R.P. (1958): Pulmonary surface tension and the mucus lining of the lungs: some theoretical considera tions. J. Appl. Physiol. 12, 262
Clements, J.A., Platzker, A.C.G., Tierney, D.F., Hobel, C.J. Creasy, R.K., Margolis, A. J., Thibeault, D.W., Tooley, W.H. and Oh, W.(1972): Assessment of the risk of respiratory distress syndrome by a rapid new test for surfactant in amniotic fluid. New Engl. J. Med. 286, 1077
Cram, S.P., Juvet, R.S. Jr. (1976): Gas chromatography. Anal. Chem. 48(5), 411R
Das, S.K., Foster, H.W., Adhikary, P.K., Mody, B.B., Bhattacharyya, D.K. (1975): Gestational variation of fatty acid composition of human amniotic fluid lipids. Obstet. Gynec. 45(4), 425 117
Dawes, G.S., (1974): Fetal circulation and breathing. Clin. Obstet. Gynec. 1, 139
Deleze, G. , Sidiropoulos, D., Paumgartner, G. (1977): Determination of bile acid concentration in human amniotic fluid for prenatal diagnosis of intestinal obstruction. Pediatrics 59(5), 647-50
Deleze, G., Paumgartner, G., Karlaganis, G., Giger, W., Reinhard, M., Sideropoulos, D. (1978): Bile acid pattern in human amniotic fluid. Eur. J. Clin. Invest. 8, 41-45
Duck-Chong, C.G. (1979): Lamellar body content of amniotic fluid: A possible index of fetal lung maturity. Ann Clin. Biochem. 16, 191
Duck-Chong, C.G., Gupta, J.M., Storey, G.N.B., Houghton, C.R.S. (1980): Lamellar body phosphoslipid content of amniotic fluid and L/S ratio compared in assessing fetal lung maturity. Clin. Chem. 26(6), 766-769
Emery, A.E.H., Eccleston, D., Scrimgeour, J.B., Johnstone, M. (1972): Amniotic fluid composition in malformations of the central nervous system. J. Obstet. Gynaec. Brit. Cwlth. 79, 154-8
Ettre, L.S. (1975): The development of gas chromatography. J. Chromatog. 112, 1-26
Fenton, E. and Nixon, D.A. (1974): The citrate concentration in adult and fetal plasma and in the amniotic fluid of the guinea pig. Biol. Neonate 24: 315-9
Gibbs, B.F., Itiaba, K., Mamer, O.A., Crawhall, J.C., Cooper, B.A.(1972): A rapid method for the analysis of urinary methylmalonic acid. Clin. Chim. Acta 38, 447-53
Gillibrand, P.N. (1969): Changes in the electrolytes, urea, and osmolality of the amniotic fluid with advancing pregnancy. J. Obstet. Gynaec. Brit. Cwlth. 76, 898
Gluck, L., Kulovich, M.V., Borer, R.C., Brenner, P., Anderson, G.G. and Spellacy, W. N. (1971): Diagnosis of the respiratory distress syn drome by amniocentesis. Amer. J. Obstet. Gynec. 109, 440
Gluck, L., Kulovich,M.V., Borer, R.c. and Keidel, W.N. (1974): The inter pretation and significance of the lecithin / sphingomyelin ratio and shake test for estimating fetal pulmonary maturity Amer. J. Obstet. Gynec. 106, 597
Gompertz, D., Goodey, P.A., Saudubray, J.M., Charpentier, C., Chignolle, A.(1974): Prenatal diagnosis of methylmalonic aciduria. Pediatr. 54, 511-3.
Gosden, C.M., Brock, D.J.H. (1977): Morphology of rapidly adhering amniotic fluid cells as an aid to the diagnosis of neural tube defects. Lancet i, 919-922 118
Gosden, C.M. and Brock, D.J.H. (1978): Combined use of alpha-fetoprotein and amniotic fluid cell morphology in early prenatal diagnosis of fetal abnormalities. J. Med. Genet. 15, 262-270
Gosden, C., Ross, A., Eason, P. (1981): Amniotic fluid cell cytology and cytogenetics. In: Amniotic fluid and its clinical significance. Sandler, M. (Ed.) (in print)
Hagenfeldt, L. , Hagenfeldt, K. (1972): Gas chromatographic-mass spectro- metric analysis of organic acids in amniotic fluid. Clin. Chim. Acta 42, 219-220
Hagenfeldt, L., Hagenfeldt, K. (1976): Individual free fatty acids in amniotic fluid and in plasma of pregnant women. Br. J. Obstet. Gynec. 83, 383-6
Hallman, M., Feldman, B., Gluck, L. (1975): The absence of phosphatidyl- glycerol in surfactant. Pediat. Res., 9, 396
Harrison, R.F. (1972): Amniotic fluid uric acid levels in the maturing fetus. J. Obstet. Gynaec. Brit. Cwlth. 79, 708-9
Heikkinen, J., Maentausta, 0., Tuimala, R., Ylostalo, P., Janne, 0., (1980): Amniotic fluid bile acids in normal and pathologic pregnancy. Obstet. Gynbec. 56(1), 60-64
Henneberg, D. , Henrichs, U., Schomburg, G. (1975): Special techniques in the combination of gas chromatography and mass spectrometry. J. Chromatog. 112, 343-352
Hertig, A.T., Rock, J, and Adams, E.C.(1956): A description of 34 human ova within the first 17 days of development. Amer. J. Anat. 98, 435
Hood, W., Blunt, V.A.W., Owen, A. (1977): Falling amniotic fluid palmitates in diabetic pregnancies. Br. J. Obstet. Gynec. 84, 824-7
Horning, M.G., Stratton, C., Nowlin, J., Wilson, A., Horning, E.C., Hill, R.M. (1973): Placental transfer of drugs. In: Fetal Pharmacology. Boreus, L. (Ed.) Raven Press, New York.
Hutchinson, D.L., Hunter, C.B., Nelsen, E.D. and Plentl, A.A. (1955): The exchange of water and electolytes in the mechanism of amniotic fluid formation and the relationship to hydramnios. Surg. Gynec. Obstet. 100, 391
Jeffcoate, T.N.A., Scott, J.S. (1959): Polyhydramnios and oligohy dramnios. Canad. Med. Ass. J., 80, 77
Jellum, E. (1977): Profiling of human body fluids in healthy and diseased states using gas chromatography and mass spectrometry, with special reference to organic acids. J. Chromatog. 143, 427-462
Jones, J.B., Pycock, C.J. (1978): Amniotic fluid levels of 5-hydroxytryp- 119 tamine and 5-hydroxyindoleacetic acid before and during labour. Brit. J. Obstet. Gynaec. 85, 530-532
Josimovich, J.B. (1978): Hormones in the amniotic fluid. In: Amniotic fluid— Research and clinical application. 2nd rev. ed. Fairweather, D.V.I. and Eskes, T.K.A.B. (Eds.) Excerpta Medica p. 209
Kovats, E. (1965): Advan. Chromatog. 1, 229
Krstulovic, A.M. (1979): Amniotic fluid uric acid levels determined by reversed-phase liquid chromatography with spectrophotometric and electrochemical detection. J. Chromatog. 164, 363-72
Kulovich, M.V., Hallman, M.B., Gluck, L. (1979): The lung profile. I. normal pregnancy. Am. J. Obstet. Gynecol 135, 57
Lavoinne, A., Loraillere, P., Maitrot, B., Duval, C., Lemoine, J.P., Gray, H. (1980): Development and comparison of lecithin/sphingomyelin and palmitic acid/stearic acid ratios in amniotic fluid. Clin. Chim. Acta 28, 102(2-3), 207
Lawson, A.M. (1975): The scope of mass spectrometry in clinical chemistry. Clin. Chem. 21(7), 803-824
Lind, T, Hytten, F.E. (1970): Relation of amniotic fluid volume to fetal weight in the first half of pregnancy. Lancet, i, 1147
Lind, T., Kendall, A., Hytten, F.E. (1972): The role of the fetus in the formatioin of amniotic fluid. J. Obstet. Gynaec. Brit. Cwlth. 79, 289
Lind, T.(1978): The biochemistry of amniotic fluid. In " Amniotic fluid- Research and Clinical Application, 2nd rev. ed. D.V.I. Fairweather and T.K.A.B. Eskes (Eds) Excerpta Medica- 1978
Loose, R., Paterson, W.G. (1966): 5-hydroxyinole acetic acid in amniotic fluid and fetal 5-hydroxytryptamine metabolism. J. Obstet. Gynec. Brit. Cwlth. 73, 647-53
Mason, R.J., Nellenbogen, J., Clements, J.A. (1976): Isolation of disat urated phosphatidylcholine with osmium tetroxide. J. Lipid Res. 17, 281-284
McCarthy T. and Saunders, P. (1978): The origin and circulation of the amniotic fluid. In: Amniotic fluid-Research and Clinical Application, 2nd rev. ed. D.V.I. Fairweather and T.K.A.B. Eskes (Eds.) Excerpta Medica, p.l
Meness, T.O., Miller, J.D., Holly, L.E. (1930): Amniography. Preliminary report. Amer. J. Roentgenol. 24, 363
Mirkin, B.L. (1973): Drug distribution in pregnancy. In: Fetal Pharma cology, Boreus, L. (Ed.) Raven Press, New York p. 1-27 120 Moore, K.L. (1973): The fetal membrane and placenta. In: The Developing Human p. 85, W.B. Saunders Co., Phil. Pa.
Morrow, G., Revsin, B., Lebowitz, J., Britt, W., Giles, H.(1977): Detec tion of errors in methylmalonyl-CoA metabolism by using amniotic fluid. Clin. Chem. 23(5), 791-5
Moya, F., Thorndike, V. (1963): Passage of drugs across the placenta. Am. J. Obstet. Gynec. 84, 1778-98
Muskiet, F.A.J., Jeuring, H.J., Nagel, G.T., Bruyn, H.W.A., Wolthers, B.G. (1978): Mass-fragmentographic determination of catecholamine metabolites in amniotic fluid and its possible clinical usefulness. Clin. Chem. 24(1), 1899-1902
Nicholls, T.M., Hahnel, R., Wilkinson, S.P. (1978): Organic acids in amniotic fluid. Clin. Chim. Acta 84, 11—17
O'Neil, G.J., Davies, I.J. , Siu, J. (1978): Palmitic/stearic ratio of amniotic fluid in diabetic and nondiabetic pregnancies and its relation ship to development of respiratory distress syndrome. Am. J. Obstet. Gynec. 132, 519
Parmley, T.H., Seeds, A.F. (1970): Fetal skin permeability to isotopic water in early pregnancy. Amer. J. Obstet. Gynec. 108, 128
Pattle, R.E. (1958): Properties, function and origin of the alveolar lining layer. Proc. Roy. Soc. B, 148, 217
Pitkin, R.M., Zwirek, S.J. (1967): Amniotic fluid creatinine. Amer. J. Obstet. Gynec. 98, 1135
Potter, E.L. (1965): Bilateral absence of ureters and kidneys. A report of 50 cases. J. Obstet. Gynaec. Brit. Cwlth, 25, 3
Queenan, J.T. (1978): Amniotic fluid proteins, amniotic fluid amino acids and their clinical significance. In: Amniotic fluid— Research and clinical application. 2nd rev. ed. Fairweather, D.V.I., Eskes, T.K.A.B., (Eds.) Excerpta Medica p. 187
Queenan, J.T., Thompson, W., Whitfield, C.R., Shah, S.I. (1972): Amniotic fluid volume in normal prgnancies. Amer. J. Obstet. Gynec. 114, 34
Raiha, N.C.R,(1963): Organic acids in fetal blood and amniotic fluid. Pediatrics. 32, 1025
Report of the U.K. Collaborative Study on Alphafetoprotein in Relation to Neural Tube Defects. (1977) Lancet i, 1323-1333
Report to the Medical Research Council by their Working Party on Amnio centesis. Brit. J. Obstet. Gynaec. 85, supplement no. 2 (1978)
Rhodes, P. (1969): The volume of liquor amnii in early pregnancy. J. 121 Obstet. Gynaec. Brit. Cwlth, 73, 23
Roboz, J. (1975): Mass spectrometry in clinical chemistry. In: Advances in clinical chemistry. Bodansky, 0., Latner, A.L. (Eds.) Academic Press, N.Y. 17, 109-191
Russell, P.T., Miller, W.J., McLain, C.R.. (1974): Palmitic acid content of amniotic fluid lecithin as an index to fetal lung maturity. Clin. Chem. 20, 1431
Schirar, A., Vielh, J.P., Alcindor, L.G., Gautray, J.P. (1975): Amniotic fluid phospholipids and fatty acids in normal pregnancies in relation to gestational age and neonatal condition. Am. J. Obstet. Gynec. 121(5), 653-663
Seale, T.W., Chan, W.Y., Shukla, J.B., Rennert, O.M. (1979): Isolation and characterization of a polyamine-peptide conjugate from human amniotic fluid. Clin. Chim. Acta 95, 461-472
Seeds, A.E., Leung, L.S., Tabor, M.W., Russell, P.T., (1979): Changes in amniotic fluid glucose, beta-hydroxybutyrate, glycerol, and lactate concentration in diabetic pregnancy. Am. J. Obstet. Gynec. 135(7), 887- 895
Seppala and Ruoslaliti, (1972): Br. Med. J. iv 769-771
Serr, D.M., Czackes, J.W., Zuckerman, H. (1963): Comparative studies on uric acid levels in amniotic fluid, fetal blood and maternal blood. Obstet. Gynec. 21(5), 551-3
Shinka, T., Hisanaga, N., Maruyama, I., Kuhara, T., Matsumoto, I. (1'978): Gas chromatography-mass spectrometric analysis of organic acids and amino- acids in amniotic fluid. Proc. Jap. Soc. Med. Mass Spec. 3, 137-142
Shinka, T., Maruyama, I. Kuhara, T., Matsumoto, I., Kabashima, K., Abe, H., Kato, T. (1979): Variations of organic acids in amniotic fluid as gestation proceeds. Proc. Jap. Soc. Med. Mass Spec. 4, 345-348
Simpson, N.E., Dallaire, L., Miller,J.R., Siminovich, L. Hamerton, J.L., Miller, J., and McKeen, C. (1976) Can. Med. Ass. J., 115, 739
Smith. A.L. and Scanlon, J. (1973): Amniotic fluid D(-)-beta-hydroxy- butyrate and the dysmasture newborn infant. Amer. J. Obstet Gynec 115, 569
Smith, C.J., Kelleher, P.C., Belanger, L., Dallaire, L. (1979): Reacti vity of amniotic fluid alpha-fetoprotein with concanavallin A in diagnosis of neural tube defects. Med. J. 1, 920-921
Stafford, M., Horning, M.G., Zlatkis, A. (1976): Profiles of volatile metabolites in body fluids. J. Chromatog. 126, 495-502
Strang, L.B. (1977): Growth and development of the lung: fetal and post natal. Ann. Rev. Physiol, 39, 253 122
Sutcliffe, R.G. (1975): The nature and origin of the soluble protein in human amniotic fluid. Biol. Rev., 50, 1
Sylvester, P.E., Hughes, D.R. (1954): Congenital absence of both kidneys, Brit. Med. J., 1, 77
Sweetman, L., Weyler, W., Shafai, T., Young, P.E., Nyhan, W.L.(1979): Prenatal diagnosis of propionic acidemia. JAMA 242(10), 1048-52
Taussig, F.J. (1927): The amniotic fluid and its quantitative variability. Amer. J. Obstet. Gynec. 14, 505
Torday, J., Carson, L., Lawson, E.E. (1979): Saturated phosphatidyl choline in amniotic fluid and prediction of the respiratory distress syndrome, New Eng. J. Med. 301(19), 1013
Tu, J.B., Wong, C.Y. (1976): Serotonin metabolism in normal and abnormal infants during the perinatal period. Biol. Neonate 29, 187-193
U.S. National Institute of Child Health and Human Development Study Group (1976): Midtrimester amniocentesis for prenatal diagnosis. J. Amer. Med. Ass. 236, 1471
Uyeno, D.(1919): The physical properties and chemical composition of human amniotic fluid. J. Biol. Chem. 37, 77
Valenti, C., Schutta, E.J. and Kehaty, T. (1969): Cytogenetic diagnosis of Down's syndrome in utro. J. Amer. Med. Ass. 207, 1513
Van Leeuwen, L., Jacoby, H., Charles,D. (1965): Exfoliative cytology of amniotic fluid. Acta Cytol 9: 442-445
Villee, C.A. (1965): Placental transfer of drugs. Annals N.Y. Acad. Sci. 123, 237-244
Voshurgh, G.J., Flexner, L.B., Cowie, D.G., Hellmann, L.H., Proctor, N.K. and Wilde, W.S. (1948): The rate of renewal in woman of the water and sodium of the amniotic fluid as determined by tracer techniques. Amer. J. Obstet. Gynec 56, 1156
Wagstaff, T.I. (1978): The estimation of pulmonary surfactant in amniotic fluid. In. Amniotic fluid-Research and clinical application. 2nd rev. ed. Fairweather, D.V.I., Eskes, T.K.A.B. (Eds.) Excerpta Medica p.347
Warren, C., Holton, J.B. and Allen, J.T. (1974): Assesment of fetal lung maturity by estimation of amniotic fluid palmitic acid. Brit. Med. J. i, 94
Weiss, R.R., Ebert,R., Dichter, J., Macri, J.N., MdcLaughlin, S.E. Lancet, 2, 578, 1979
Williams, J.L., Bargen, J.A. (1924): The uric acid content of human amniotic fluid. Am. J. Obstet. Gynec. 7, 406 123
Williams, V.P.; Ching, D.K.; Cederbaum, S.D.(1979): Adsorption of organic acids from amniotic fluid and urine onto silica gel before analysis by gas chromatography and combined gas chromatography / mass spectrometry. Clin. Chem; 25(10), 1814-20
Wilkinson, S.P., Niven, S.E., Hahnel, R.(1979): Quantitation of organic acids in amniotic fluid by gas chromatography. Clin. Chim. Acta 99, 241-5
Zuspan, F.P., Singh, E.J. and Mejca, A.(1975): Palmitic acid / stearic acid ratio in'lecithin and fetal maturity. Amer. J. Obstet. Gynec. 121(4), 577
Zuspan, F.P., Behrman,R.and Paton, J. (1974): Amniotic fluid epinephrine and norepinephrine I. Comparision between the human and baboon. Amer. J. Otstet. Gynec. 118(6), 837
Zambotti, F., Blau, K., King, G. S., Campbell, S., and Sandler, M. (1975): Monoamine metabolites and related compounds in human amniotic fluid: Assay by gas chromatography and gas chromatography-mass spectrometry. Clin. Chim. Acta, 61, 247-56 APPENDICES
Appendix A :
Compilation of capillary GC-MS profiles of basic fractions of amniotic fluid, serum and ovarian cyst fluid.
124 TI r— I— i— T
Fig. 24 TI plot of trimethylsilylated CH2CI2 extract of basic components of AF (15.5 weeks, tapped for maternal age). 125 Fig. 25 TI Plot of trimethylsilylated CH2CI2 extract of basic components of AF (16 weeks, tapped for maternal age). 126 TI
Fig. 26 TI plot of trimethylsilylated CH2CI2 extract of basic components of Af (IUGR, inderal, apresaline, hydrochlorothiazide, hyperplasia of renal artery). Fig. 27 TI Plot of trimethylsilylated CH2CI2 extract of basic components of AF (open tubular neural defect). 128 Fig. 28 TI plot of trimethylsilylated CH2C12 extract of basic components of AF (30 weeks, severe Rh). Fig. 29 TI plot of trimethylsilylated CH2CI2 extract of basic components of AF (37 weeks, Rh sensitized, OD450 = 0.01). TI
Fig. 3Q TI plot of trimethylsilylated CH2CI2 extract of basic components of AF (16-18 weeks, nicotine 0.5 pack/day, Tylenol, vitamins). Pig. 31 TI plot of trimethylsilylated CH2CI2 extract of basic components of AF (38 weeks, Delalutin once/week, tapped for L/S). J i2_.i3 .n is i6.r;:-.-L3-La_ga..^i aa ^3 ai as 26 a? S3 a9 31 aa jj 34 as j6 J7
Fig. 32 TI plot of trimethylsilylated CH2C12 extract of basic components of AF ( hydrops) • fa__ L — ^ ^ " i *t— I— t ■ i— r— i— i— i— i— r— i— t— i I i— i r t i i i r i i l a 13 14 15 16 17 li 19 d3- jS-JZ.
Fig. 33 TI plot of trimethylsilylated CH2CI2 extract of basic components of AF (diabetic). 134 Fig. 34 TI plot of trimethylsilylated CH2CI2 extract of basic components of AF (Down syndrome and GI block). TI ■' I " r I I '1!' ! " 1! I 1 I' I ' I » 1 I Y f ! I I I I i l l i d I :a i ia n ia 13 n is 16 17 18 19 d*d cil cia_c:3_dj-g5_Sa-S7Lga-ga-3
TI plot of trimethylsilylated CH2CI2 extract of basic components of AF (drug addict). Fig. 36 TI plot of triraethylsilylated CH2CI2 extract of basic components of serum (open tubular neural defect). 137 Fig* 37 TI plot of trimethylsilylated CH2CI2 extract of basic components of ovarian cyst fluid. 138 Appendix B
Compilation of mass spectra.
139 LACTIC ACID DI-TflS 2b5&
50-
tim HH|Hii*iUlftnT 4 n r | t f l^ t f r >H iirtH q -ii8r » n |>h n n r n |iin ti>M |Tm~»»il|lih *iin|irir i m jn ir rnij>irrirnjiiin‘» n i p n m t T 40 90 140 190
1 0 0 —i
50- - 15 M
\ kU n iiipMi nn|inr'rnljinr ini|inrnM jnrr nnpm iiiijim nnpm iirTjm r nn|un'nii|m r inijim nrrpin m i|rrn-nTT| 190 340 390 340 (m /e ) amu
Fig. 38. mass spectrum of lactic acid di-TMS 1 0 0
*abuac rei 2-OH-BUTYRIC ACID DI-TMS
50-
rrrrr"rtirjmT-ntipiii-nitpilrniipn r n n p H rintjInr m T p u i'ltnp m nitj 11TTTTJrjnfr fjmr *1 4 0 9 0 140 190
X re abun
fl - I S
4hTiinpui‘liiipnrrht|litr inipi t hinptn nnpm itirpnruupTnirnpr 190 340 (m /e ) amu
Fig. 39. mass spectrum of 2-OH-butyric acid di-TMS 100
?b5Si 3-OH-BUTYRIC ACID DI-TPIS 50-
mu‘uiijm rtnlpin intfitllMtujmrrrlipHr mlpih'miim riHljnTruHlntrnrjhlr r rtTjrilh iin jnit iitf|nn nilpnrfllipnrnU^ 40 90 140 190
1 0 0 -i
% re I abund + n -15 50-
’
nllr-frrrniibinrm n-rriiiim inrrttKm nin nrim rrm m rtnm niiii m innrnm nn m nnn mmm muiMi-rimrni mn 190 240 290 340 (m /e ) amu
Fig. 40. mass spectrum of 3-0H-butyric acid di-TMS 1 0 0
V0 rel abund 3-0H-IS0BUTVRIC DI-TMS
SQ>-
rrrn rlnprrrtn ijlui im jitW m tjtm m tjm i 't»m filr’imi n m HlttjiTTi•inl|lilg,*iin|illi JlJtur n1t|n rr iiTilini iHirrmr 40 14090 190
100
*4. re
n +- i 5 9 0 -J /
litI f TTrTjintMtiTp n r-itlljtin itirpjn ill n*rm |m T~im |nnrinipnTTiTrjiiM n»i|m im m iii m i|nir im piir rni|HM rm|nT»~Tin^ 190 S40 S90 340 Cm/©) amu Fig. 41. mass spectrum of 3-OH-isobutyric acid di-TMS i e c
^ rel 2-0H-2-NETHYLBUTVRIC-DI-TNS abuncl
5 0 - M+ “ 15 / 11iVrpVrfjilYijliWjfrrrpiWjWrrpYii*|i jnjMrrrjiV 1/jYt*hpvYi-pViiji11>| irtjiVirprii Jp 111 p 111 |ViiVji1frpriV| rm jrmj 40 ' 90 140 190 240 290
100 — •
50-
tTrTiyn «,i‘|mr|TTmrj-iTn|ii ii|M*ir|m Tprn[ii m m r|iinjn njiiiinni|rrnpiii|ii ii|HM|im nin|in i|mTp n inm | 290 340 390 440 490 540 (m/e) amu Fig. 42. mass spectrum of 2-OH-2-methylbutyric acid di-TMS i © 0 ..
% r e t d b u n a ACETOACETIC ACID DI-TMS
5 0 - n
> titi 111 m j n n p n r| in rj i ri ij n n j ii 11 |Yi 111 l'|T 11V'VTTT1 | I I 11 I I I II IT T |‘l |»H nm I |*l|'i iII iil'| j rII u 1 Ii j'l|‘| iIT t iI j rrririftTTTl*r! 50 100 150 200 250
1 C“? ACETQACETIC ACID DI-TMS % ret, afo unci
5 0 - M
nVrpiTrpTn’ji trr j n' r/jn‘i‘1 rjvWrj 111 i‘|'ti‘n|liVi jim |iVrT|lVrryrrr urn'irn-i W i | lTtrp i frp i f Tjn i n Vi r ry 40 90 140 190 240 290
Fig. 43. mass spectrum of acetoacetic acid di-TMS abund l e r ^
bun0>2Z re 0 0 1 10® 10® Fig. 44. mass spectrum of benzoic acid TMS acid benzoic of spectrum mass 44. Fig. -j 0 5 A S:© 9 30 9 40 9 540 490 440 390 340 290
0 4 —1 —I mrrprrrrpTrrprripinpiTr|iiiYpiii|ii rr|Tr|nn n rr| rrrr|iT | n n , ,ll BENZOIC j ______11111 0140 90 it n 111 upnTpi Tpiijnipnvpinpmp i p i m | i i n ~ j m pi p n m i p p v n n iip iij n iTTpi p in p T n p u CD TMS ACID i i p i i i p i i i p i i i p i i i p i t i * .i Ii Cm/e) amu ...... I 11 Pi j 190 1^*111 N + i t * i ip i V r j n i t p i 240 11 11
jii iiprnpiiipilipnn Jvi r t | r i “ rr p i npi 11
pi 290 i> j SWI-TP Eaan jo uinajoads sseui ‘gi? *875
niu« (8/ui j 0*S 06fr 0trfr 060 ot^e 062 Linlnillllll Imiln it! mi iiAxltmiiiJilimJinilimliiulnnlimlnnlnnlm iln nln ulii 1111 ^ ■ l,i i i J 11.■
t U -03 K)
"-001
062 0tr2 06 T 0tr% 06 lnyjnn^mlii 111«11111 n 11» j(i .J n 1,^1 j i « l^llI^111 j^lj 111 l.l.ljj.,1,1,! 1^1^ u.il I4J I U,l,t,{j.l.l,l.ljj| ll j
“ 0 3
mtnqe sui-ia yaan %
0 0 T 100- 2-KET0-IS0CAPR0IC ACID DI-TMS % r s l ebuna
5 0 -
iiVi-jiiTfjnn*ji TrjiiTfjm rpritjiVrtji i[Wit|irfrrpik ill ifprn 1pirijirrn111prn11mpmj*ripT n p m y 11 ITp 111111*11111r^T 11 40 90 140 190 240 290
100
SbCSd -
50-
iiiii[iiii[nn|iiiijiiiiji m | m i|Trnj-nii|ini|'rm| iTTr|mi|iTiTfim jTiiifinT|Tm-jm i|iiin inijMii|iM i|iiirjTrrr| 290 340 390 440 490 540 ( m /e ) amu
Fig. 46. mass spectrum of 2-keto-isocaproic acid di-TMS 1 0 0 2-0H-CAPR0IC ACID DI-TNS % r e I acund
5 0 -j
TTTTT »iiiii*mijiiir imnii iTTTJTTtT 1}|mr TTT w •HTtjltn*nhjn»*»tn 40 90 140 190
100 -j i b S t i -
50-
n+ - i 5 i ,im ittljn l r m rp r ir m l j nrr-TTTT|nlt*im jiin *m i | i n i*n itj ltnr-n u |H ir »in j i ni •in i ji i ir i i n j m t i m jT»rr in T|n n im u m ' m if 190 240 290 340 (m/e) amu Fig. 47. mass spectrum of 2-OH-caproic acid di-TMS 109-, 2-KETO-UALERIC ACID DI-THS S'* re I a bu n d
5 0 - n
i iTfnrrrifrprrfj'n n r«*prr Inr ji frrjiirrjniTr jiVl^rj iTi,itjMrrrjt*i*i^fpi rfrjn ri| rrrrj i n fpVrt|Vrrtp i*iV|Y> 11 p nTjti'h j i i ffp'i 1111 in ji 11 rj 4 0 140 190 240 29090
1 0 0 -i abunct
50-]
rn ti | n rrj i u 11111 i j 11111»n i j 1111 j n 111111 i j 111111 rn j rn i j rm | «n i j n n |r rt rj vri ij 1111 |n h j 11 n j n m | m ti pTTr|TTTT|Ti i q 290 340 390 440 490 540 Cm/®) amu
Fig. 48. mass spectrum of 2-keto-valeric acid di-TMS 1 0 0 2-KET0-IS0VALERIC ACID DI-TMS % r e I, abuna
50 M J n^riTrrfjiiTi^ll^rjTiTrjnTTpv ir p W rj11 i[Ti-Jp iTrpn r jm irjnirpn i|Vr r r | r p n ijn ^|m J |>iTiyrhiy 40 90 140 190 240 290
1 0 0 i
% re I abund
504
rnnjn ii|nn|iiii|ii i i|t iiij h| iii injn iijim ju n n i n|Trri|inrpTrfjnH|ii ii[ii npi ii|imyiTnjTnTj-MiijTm|iriT| 290 340 390 440 490 540 (m/e) amu Fig. 49. mass spectrum of 2-keto-isovaleric acid di-TMS 1 0 0 1 r e I abund OCTANOIC ACID
5 0 - + n - i s B.G.
JUL|lll'rf|'nhl|
100
*4, re
50-
»i iti ii 111»rn | n 1in titti 11111 iit i r uin in h i t Mj m i | ii n jn rijii »i|ii i>|in i|H Trp rii|H Ti|fi>t|«>TT| 290 340 440 490 (m/'e) am u Fig. 50. mass spectrum of octanoic acid TMS 1 0 0 -,
abun PHENVLACETIC ACID TP1S
50-
TtiPtn Ij TTT TTT 40 90 140 190
100
% r e aburn
50-
u mmi imi iiu mrmnnii | n u ~n n jrrm m i|TK T*TTn ji H f tfrr 190 (m/o) a m u Fig. 51. mass spectrum of phenylacetic acid TMS 1 0 0 -
% r 1 1 abun TTT rr[ i rrrjm rpl iViji»rrpn i |rrri j i »‘i»rpTrr|sVi*i |TrJf|Vn ij riYrp rtt|Vi t fr^Wrn'iVrjVi 11 p »Vi ji i i i-j tn i j irr■ 11 nrrj ^ m pVi i| 40 90 140 190 240 290 100 2bSSi 50- n rrrrfJlh rrjillhjriiiiirrTjTin |in in m [rni|mi|M H fn 'iTpT n |r m jiim THTjmi|Trrrnin|iTiT|nn|mrji»»>j»»**|i>»»| 290 340 390 440 490 540 (m/e) amu Fig. 52. mass spectrum of phosphoric acid tri-TMS 1 0 0 - Js ret abund 2-KET0-3-METHVL-UALERIC ACID DI-TMS 50 nriJmTyn ii-in>|HH*in l[H tShT^n tr iriijiiir inijrm u ii|» h ‘iltntm d h jliir tm jln i 40 90 140 190 100 n abuntd 50- n rntTrrrjttirm TprriJtnr|irn~nrTtjH % re 2-KET0-CAPR0IC ACID DI-TflS abuna 5 0 -I tm r l m p1111AttiIttIit itrrn h i^iiritpnf-lnvpirrm m tpmpiirn trpm impit» nnpm Im prn-mTj 50 100 150 200 100 re I abunid 50 n tTTT--iilTjnri^rrrpiir rrrtjHn iiiijrtn'iiiipin-rtilpiiilinnnrml1 im pn rH frpfn'iuipiir unpiri nnpm rnijn ii im pin"nnpnniir| 200 250 300 350 (m/e) amu Fig. 54. mass spectrum of 2-keto-caproic acid di-TMS 100 %> re\ N0NAN0IC ACID TMS a bund n +-i5 / I B ♦ G » r J j J j V l t y W l # f | i L |W ff[W rt| J t| YtWjlhnrjl^YljWf^Ti«>|vrf/|iiY ^ f n |S i fojfl IfjM i rpi'hi | ii a 11 n n ] i inI II | I Ml |'l I I I | 40 90 140 190 240 290 100 X re l abund 50-1 iTi,r|iiii^rrn |ii4t|Tiri|‘nrrjTrn |ii n |im jn ti|im jm rjn n j n m ii 111 r rti i111 ijin ijn rijii ii|m T|rirryrm [n iijirirj 290 340 390 440 490 540 (m/e) amu Fig. 55. mass spectrum of nonanoic acid TMS 100- DECANOIC ACID TMS 5 0 - M rrn 1 lf|Vll^ i‘iVfjlnliT|Tt,ifyrm | WrjmP|iTrt|'n n|\rrr|i»W|1i rrjirtr\ij iTr»jnrr|TTTT|mw f|irrT|mi i j i jin rr p T 77]TT 40 90 140 190 240 E90 100 50- tTTTrjm ijn iijii iijrri ijii ii jim |i>n |in i|n~irjiTn|Tntjn-irjrrrrjnTr|iTTi jiTTrprm j-m tfiTnjitn |ii iijii m w n jm ij 290 340 390 440 490 5 4 0 (m /e) amu Fig. 56. mass spectrum of decanoic acid TMS i e e - k re\ PHENYL-dg “NANDELIC ACID DI-TNS a buna 50- n+“ 15 \ ■ ■lri|iirf^iiTrp^i>rp>TTi|iVrrpi‘iY|rn‘rpltrjN4TrpT»lt|i‘fr>jm lfprir|ti^rjitnrmr p injM rfjiTiniiii|rin|iiilfjmTyn iijnli| 40 90 140 190 240 290 1 0 0 ^ % re I abuna 50- nf.j|..«i|..,,|..,f|ininiT(|iiit|iiii|m i |ufr|iifin nijim j iin |niijiiirp m p n i|iiiijfnn nTi|rnnt?np>iipi)r| 290 340 3 9 0 4 4 0 490 540 (m/e) amu Fig. 57. mass spectrum of phenyl-d5_mandelic acid di-TMS 1 0 0 - +-15 SALICVLIC ACID DI-TMS M % rel abund \ 5 0 - i*rrj iTiT|-»iTrp ViYpmr |Trrijn ‘fij i rrrpii-i i lylyrtTi A ^ i jj iTm ^ Tp f V ifi'p i l r p n iu 111 p iVlYfjTi ^ jt i 1111 }[ir iW n i | 1111 rrrjl Jj 111 prrri j t*n iyi r i i/ji i n prrrpTirp i n y 40 90 140 190 840 290 100 — *4 r e I abund 5 0 4 i ii i i |i irr|i m p r>r|M ii|)f irjTu iyrnij iriTj ii n jiT r rjrTii ji i ii|iiii|i i ii| i m |n u jim jm i jim r jri iijii iijii i ij n rrjr n r | 290 340 390 440 490 540 (m/e) amu Fig. 58. mass spectrum of salicylic acid di-TMS 1 0 0 - PHENYLLACTIC ACID DI-TNS °4 ra I abund 5G- L I TI Vri'nYfjiYn'jl i jf|VnvjlH*fqiSliTjT»l^piW |^lr>pYi lprrfftWrpvft| rrrrf rp ‘fr|rm Vi'iTjIT|11111 i n »piTi pTTf | tt p riTrj*rrrrj 40 90 140 190 240 290 1 0 0 -i X r e I abund 50- i+- 15 TTTi/lT|in i|im |nirjnTrpT»T|iin p Ttp ii|,rrfrpiiT | m i[iin|iin|iTTijim n r n |HM|im|»nnnii|iiTTjmiji»u|»iir| 290 340 390 440 490 540 Cm/e) amu Fig. 59. mass spectrum of phenyllactic acid di-TMS 1 0 0 n 4-ACETYLAniNOPHENOL-DI-TnS 5 © h Wfq WrtfH VrrjtfifrrffoAWjIvi /|Wrrp ^ Lrfj^ii ijllfrfjil l*pirrJ|T¥iijn^Pj>rnptfrjifmrrrrt Ivrrj 40 90 140 190 240 290 100 abun t»rninirpirn rriiff»njin ip nT|iim irii|rrii|iiiniin|ii npnijii 111 |~pTrT| 11 n | T n 111h t j i r i »11 n i'| r m i~n TT| r f 290 340 390 440 490 Cm/©) »mu Fig. 60. mass spectrum of 4-acetylaminophenol di-TMS 100- b ( & P-OH-PHENYLACETIC ACID DI-TfIS SO- TTTTT TtTtfrtn -niip ih itii|it Hmnitn i y r t n i m u j t m m !* 1111 V w TT 40 90 140 190 100 50- nni‘Hnjnii-m >[iill m ljtiif‘Htijnn tllt|un*ltH| tii[iiii~liiijHTT'itn|HI,i-it*m in~ Illfju rT riTrjin rth ijim m ljriU Tnq 190 240 290 (m/e) amu Fig. 61. mass spectrum of p-OH-phenylacetic acid di-TMS LAURIC ACID TMS h re I abund 40 90 140 190 240 100 — *4 r e abun 290 340 390 490 (m/o) amu Fig. 62. _ mass spectrum of lauric acid TMS 1 DO-, H0M0UANILLIC ACID DI-THS 90 140 190 240 890 50- TTT TTTTT Tn ijTM »jn ii|Mnj~n m n TTj rrr» p i i i j i i ii j 11 h j 11 n j i it ij u 111 irrrjn 111 n »rj»i n j rm | r i". 290 340 390 440 490 (m /a) amu Fig. 63. mass spectrum of homovanillic acid di-TMS 100- HIPPURIC ACID DI-TMS tyb re I, a bund 50- rmfcrjltWp vtfft bp i*fr|v»W|HVrrpVi^jlvrrp1!’!^jrt'pr t f p nTj& irphk'piiipifrpirn m rpu rf 40 90 140 190 240 290 100 % re sbun 50- ii 11'| n i tjirfrjrn i jin ij m i'irrn p irrTTTTi' 111M 111111II111iTfrm Tnm im Tr p m p i n jn ii p » n j 340 440 490 540 (m/e) amu Fig. 64. mass spectrum of hippuric acid di-TMS 166 100-, HIPPURIC ACID TMS % r e I e-bund n iiyi|W rpiirpW rp iiTpTMpv}Tpiftp iiip1'ripTrt‘pm pViip iilrpTn|t tiip11 p i iltpSt*| i n m i p i H i i ip j 11 iripm ii 11 n i |V |lh‘frpi ripiiip nr, 40 90 140 190 240 290 100 - % re I aound 50- in ir|T r »fp H irp irrp i i ip iirjTTTrp-rTip in y n ,»»|n T r p i ti|iiiip u i|TTTr|'n»i p irip n ip n T |n n i T r r « y m » j n n j n u ^ i 290 340 390 440 490 540 Cm/©) amu Fig. 65. mass spectrum of hippuric acid TMS i 00 % f e l a b u nd P1YRISTIC ACID TMS 50- »iMrnVl I /fj^lIY|tt*ilfjWnljV>*rlpYri jntrj'A Itjtrf fyvrrjVi ri p iVrpTi irFj iVrrpirfjii rrjWrrj h^r 40 90 140 190 240 290 H r e abuni 50- t h i pnnrr|»i n |Tri'ip m | n fr p m |iTrrjnniTiinnii|rn T|»m Y r r m r m y rrinriTi|Tmju n |n nin ri|nii|iMT| 340 390 440 490 540 Cm/e) amu Fig. 66. mass spectrum of myristic acid TMS 1 00 -1 % re I abund UANILLYLPROPIONIC ACID DI-TNS 5 0 - II i foi i jjn*! l*i ^ iVnjWn jlA ip Iri ji^ j i ri 100 abund\ r e Jf 50- n Irrf ir|i m j irrip T4TjTm | 4 rii | m ifirrT[>iii|im | m r p Tii|M ir|TTn | m T |n ti|n TTfn tr^TTrrp->ir^»rrr|n ii'jirn p TirjTrrrj 290 340 390 440 490 540 (m /e) amu Fig. 67. mass spectrum of vanillylpropionic acid di-TMS 169 I©©-, ■ IND0LE-3-ACETIC ACID DI-THS 5 0 - r mrrprr r j i rrrp m p iT^pVrrp i*rrpm)r|irr JjlYrr jiivf pTrrjn rrjirrrj i rht| ti 11 fT| r n r pnu r|T> r ip m 11 rrrp i n p i frrp rM-| 40 90 140 190 240 290 1 0 0 -i % re I abund 50-1 n In u 'p n rfin Vrrri11 iirjm i j i it t| ruim iniiiijii m rrii|ii in in i[ii rr]iinim |irir[iin riiiiiim iiiinj irri | mi[ 11 ITjllTqllllj 290 340 390 440 490 540 (m/e) amu Fig. 68. mass spectrum of indole-3-acetic acid di-TMS abund % ■i bund re i. 69. Fig. re I 100 0 0 1 50- I 3 y •• y 3 9 30 9 40 9 540 490 440 390 340 290 0 9 10 9 20 290 240 190 140 90 ' 40 Wj^rp i r Tpr i pmptitpiiipMrTrp Fi npMi pMiprT ' pmp-iirj p m ip p'M m p Tr r p i M ip i i ip M p in ip iF p i rjTTrrpn M p i i i p t i t p m ip M p iiy p i ri p iT i p n T rrp r p 'ir rrpt js^ nW +-15 M mass spectrum of pentadecanoic acid TMS acid pentadecanoic of spectrum mass m/) amu /e) (m ETDCNI AI TMS ACID PENTADECANOIC tee-, Yc r e l a b u . n a PALMITOLEIC TP1S 50- f|livr|Vrjffrt‘iT|Vi,rtlpi>^r|ln,, jSr, i j, i fr| 11 n ft m j 190 240 290 100 - , *‘e t buna M 50- m q iii 11*11rrr 1*1 iijiih j»i in n m » i‘u|irii|»m |»rn|innninim|iri»jn n|Hlijn■»[»■ '^|*» »»|»»I.ip...ji.nj 290 340 390 440 490 540 (m/e) amu Fig. 70. mass spectrum of palmitoleic acid TMS 100-, % re I PALniTIC ACID TMS aJbund 50 ti'iYrpl■JlVp n V jl I I VtrpYPrprrrtji'r»ij^TrpiVrpV r r r|i'llipiiYjn up1* nprrt p iiVpiii pVri| 40 90 140 190 240 290 100-1 % re I abund 50- PI til I CjII ITJ'I iVip n fyn trpTu p rrrprn pi npTTrp TiipTrii m Tp Tirprn-pTtipTiTpmpriipm prrrpmpi np» npuTj 290 340 390 440 490 540 (m/e) amu Fig. 71. mass spectrum of palmitic acid TMS LINOLEIC TmS lt rrpvrrpvn j lVrr|i11 rp r i i j 40 90 140 190 340 290 100 - i r t t n i i i f| ii 11 |~ni‘i | r r r F jti 1 1 1 Iv nj i i 11 111 r r p i r i| ii iiii i n |m r jrM rjim j iriinnT|rni|r m p M r|m T jn n ,ji,nT|iin|THTj 340 390 440 ( m / ' o ) a m u Fig. 72. mass spectrum of linoleic TMS OLEIC ACID TPIS 40 90 199 100 * rel abund 50- M Trri*t i 111 i'iVi 111 Vi' *n n |*bT| im | iTiTji i iijmiji i ii j nn | n ii[Tni|» ■ t ■ | »> »[«■«i|ii «»| ■» »■[■ ■■ ■ | ii^n')Tj 290 340 390 440 490 540 (m/©) amu Fig. 73. mass spectrum of oleic acid TMS 175 STEARIC ACID TMS 40 90 140 190 240 100 ItXSk 50- TTI j i T rfj iTiT|~r rn jm i1 jn i 111 n i jr m j m 11 u 11'[ 340 390 440 490 (m/e) amu Fig. 74. mass spectrum of stearic acid TMS •^J O' SQUALENE (?) 40 90 140 190 100-i %. re I abunid 50-1 iiW |i»i^i'iii|ii,i,rin ii‘ll»iii|ii,ri|ni,rn,>in m H iinrrr m »i,»»iiiM,| iV ^pi-ffp Vi rpI‘/| | 1A i’p »I »p T r rp T r r j 240 290 40 90 140 190 100 - *4 re I abund pi 11 pTn 11 rf f 1111 rj 11 iTp^h 111111| i ■ i ig 11 ■«{ itii p iTrj > 11111 n i( n i ij 111 ^ »p "' | • t •11 • 11'' ■ |] •111'1 *11 ri~^ 340 390 440 490 5^ (m^e) smu Fig. 76. mass spectrum of cholesterol TMS 178 CAFFEINE 50- rm i jT^h jt rfrj rrrrp i 11 j it n |»i rfji nij IT 7T TTTT ttji n Tj m i | i n ij»it r j 11111 n 11 j 90 140 19040 240 290 Fig. 77. mass spectrum of caffeine 100 COTININE coO- IfTTTini TTTtjT fTtT 40 90 Fig. 78. mass spectrum of cotinine