GLC CHART RECORD OF TFA - OIPEPTICE METHYL ESTERS TOTAL TON CURAEAT 4. Y T u Y T T T eee T T T TeVPry T 1 ad we Bum are vee 208 aes 048 vere cm msi is TIE Syme 6.36 HIN, A COMPUTER PLOT OF THE SUMMED PEAKS (IONS) OF EACH SPECTRUM Figure 5, The above trace is a GLC flame ionization chart of an expcriment. Bclow is a total ion plot, created, and plotted by the computer, that indexes the individual spcctra OPTION =?TYPE EXPERIMENT? © 31'S26_15 FROM SPECTRUM | ©2555 TO =7960 BY MASS, ANP, of ENTIRE =2ANP FROA MASS = © 230 TO =?750 NUMBER =29 SPECTRUM 955 MASS 31 «62 &3 46 $7 69 139 366 16? AMPL 203° 22 22 #26 22 #930 23 37) «02? SPECTRUN 956 WASS 31 4S 'E 69 70 $6 139 166 1857 AMPL 117-250 2832S CKO SPECTRUM 957 MASS 3104S) OMS) 697071 OT AMPL 1246 38 39 5? be 52 &? 102 76 SPECTRUM 958 . ; MASS 32 &3S 65 69 ‘270 72 199 166 167 AMPL 115 32 32 36 31 3H) 32 60 OE SPECTRUN 959 MASS 31 63 56 69 70 S98 439 166 167 AMPL 116 26 26 26 27 25 29 © «©6§1 © 639 SPECTPUN 960 wASS) 3100 32S KCK AMPL 1139 22 246 23 25 2 21 36 30 Figure 6. The principat peak listing from the matrix search times the circuit transfer function over the frequency range in w. The first shown below (on the left side of the equation) represents the white noise retained by a full integrator during a time interval 7. The second eéxpresses the white noise passed by a simple amplifier of bandpass, fig = 1/(2 wreq). CJ T 2 =) | 1 f f ec dt| dw = f \e" —_—_—— 0 0 0 1+ jute A solution for req may be made by numerical methods: Teq == 0.496 T : (4) Using this value, it is found in the example that f,, = 40 Hz. A last analytical benefit of the system is the efficiency of detccting ions. In any MS with electronic ion detection, an ion is detected or not detected and there is no “5% detection.” dw = (3) There is a 5% probability an ion gencrated in the source will pass the analyzer and hit the detector. Or more accuralily, if when a peak is “centered,” n ions register upon the detector each millisecond, it may be expected that at the appropriate low side of the peak, n/20 ions will register each millisecond. The contro! method is always “centered” upon the peak, hence the total expected ions will be T X # ions. However, in scanning a gaussian peak, it can be shown that for an cquiv- alent time, from the first 5% point through the maximum and out to the last § % point, the total expected ions will be approx- imately 0.51 X 7 X 4, where now T is the time from 57 point to5% point. These numbers ofions expected to register upon the detector are tabulaied as “Ions Detected" in Table II. Many small, but important, peaks will have an # of 1 to 10 ions per millisecond under the operating conditions imposed by some modern research requirements. A figure of merit for comparing the systems may be defined: 1000 Detected Ions —_— (3) Bandwidth (Hz) Figure of Merit = This is the fina] tabulation in Table Hf. Our use of the system has verified these results, that introducing computer control to the quadrupole or T-o-F MS can enhance the useful sensitivity by a factor of 310 8. The grcater convenience offered to the user is not at the expense of performance, but olfers these intrinsic advantages. One other practical benefit is possible by reviewing the attributes of resolution. The idealized gaussian peak con- sidered here, which has a width commensurate with the 107% valley resolution criteria, has a contribution of only 0.06° at the neighboring integer mass position. Since in the control method, only the cross contribution at the integer positions is harmful for some experiments, the resolution mey be de- graded until a 0.5% or 1% cross contribution is observed. If the physics of the particular instrument aré exploited. it will be found that this will greatly increase the portion of gener- ated ions that will be passed by the analyzer. Some consideration should be given here to the software and its interaction with the user and the instrument. A mode of conversation has been programmed into the system to prompt the user for necessary parameters, and then expand upon thesc parameters to conduct the necessary detailed operation of the instrument. In retrospect we find that we have very closely followed the concept of R. J. Spimard (2/). A simian problem occurs with the presentation of information. which is often too voluminous for complete display. To overcome this, the bulk of the information is held in the computer. and is made available by similar conversational techniques. It should be realized that the present software is in an cm- bryonic state of development compared to its eventual poten- tial. The system of computer, interface, and MS may be defined in terms of computer syntax to a systems developer in a manner comparable to the definition of a specific data process- ing task to a system programmer. There is an enormous po- tential in the implemented syntax of this instrumentation system for the chemist who may have special requirements and who understands fully the principles of the augmented instru- ment. The user con then program, or have programmed, efficient solutions to many of his instrumentation problems. The features that do allow the foregoing benefits, do them: selves impose certain limitations. The control system docs not normally return any information about doubly ionized odd (21) R. J. Spinard, Scicnee, 158, 3797, $5 (1967). P-7% 1128 © ANALYTICAL CHEMISTRY, VOL. 42, NO. 11, SEPTEMBER 1970 CHj;-—— cH, CH, —— CH, CRs f's | | | | ¢ 166 % , & pe CONH—~CH— COOCH, 4 nt | | i. Cm CF, cry 4 cH, — i cH, Le caot , i C=—~0 ; | F (m-och,]+ = wt | s4 3 293 324 | HH HD 20 4 60 80 190 120 140 160 TFA DIPEPTIOE METHYL ESTERS HS26 15 FILE 957 1800-200 228 248 260 280 320 329 349 Figure 7. The on-line, computer plotted spectrum mass values (peaks at 0.5 integer position), wide metastable peaks, or the value of mass defects. It also might be expected that mass defect deviations, especially at high mie values, would cause difficulty. In practice there have not been any problems attributed to this latter case, the high wie defect. The resolution and other operational factors have masked this inherent difficulty. If operational use should be hampered by this problem, we have proposed techniques of concurrent calibration on the unknown peaks themselves. For example, the major peaks in the unknown compound, identified to their nominal integer value (less mass defects), would be used to calibrate the N table. After a programmed computer decision to explore a peak area, similar to those illustrated in Figure 3, the computer would direct the “scan” to that area. Data acquisition time would be about 100 msec for 10 data points over the peak protile. Centroid identification and recalibration might be another 100 msec. If this were repeated for 8 or 10 key peaks, total recalibration would be in the order of 2 sec. This technique could also be employed to verify the linearity above the last reference peak (614 with our system) or below the first. (We have paid little attention to m/e below 12 or 18.) As a final minimal operating mode, the scan could be pro- grammed to simulate any present mode of time-based scan and data collection, with comparable attributes of performance. As an example, doubly ionized masses or metastable peaks might be identified and measured. Of course the MS used must have sufficient resolution and/or sensitivity, also the total spectrum acquisition time would be increased. Such changes of operational mode would be by program only; hence they could be established or removed in milliseconds with no hardware or switch changes of any nature. RESULTS In a typical experiment a mixture of TFA-dipeptide methyl! esters (22) was injected into the GLC-MS system, and 130 complete mass spectra were collected and recorded by the computer. Eighty-one of the individual spectra were summed to obtain the total ion plot (Figure 5) (/¥). This may be com- pared with the GLC flame ionization chart record on the top. The mass spectral output from peak “C."" for example, in the (22) B. Halpern and J. W. Westley, Biochem. Biophys. Res. Conmimuur., 19, 361 (1965). ANALYTICAL CHEMISTRY, VOL. 42, NO. 11, SEPTEMBER 1970 e region of scans 955 to 960, was then abstracted by the TYPE routine and the results are shown in Figure 6. This data shows that Peak C is homogeneous and that a satisfactory normalized plot can be obtained from scan No. 957. Figure 7 is the plot as produced in the laborato.y. The background in the mass spectrum is primarily due to the GLC column bleed. Only the chemical notation was added by hand. The taking of spectra, in this computer-compatible form, opens the way to many types of further processing: searching through each stored spectrum to pick out the amplitude at a specified mass position and plot this data to show the variation of the ion current of a specific nz;e with time; check the homo- geneity of each GLC peak; resolve the mass spectra of simple mixtures into those of the individual components; and sub- tract background signal from a mass spectrum. The system is also well suited to the recording of spectra from solid samples introduced directly into the ion source. Because of the fast scanning speed and the recording of many spectra, there is no need to establish a constant vapor pressure in the mass spectrometer and any contamination or decom- position of the sample is readily detected. * CONCLUSION It is feasible to build a limited purpose computer and a quadrupole MS as one unit, forming a compact instrument with features similar to the system presented in this repert. In addition, some of the software could be embodied in hard- ware circuits. Such units would be more costly than those using commercial general purpose computers, but could be uniquely well adapted to space or other special physical en- vironments. Alternatively, the system can be interfaced with a larger time-shared computer. RECEIVED for review March 23, 1970. Accepted June 15, 1970. This research was aided by grants from the National Institutes of Health, Public Health Service, AM 12797-01 and FR-00311-04. The LINC computer was provided under NIH grant FR-00151-01. The work is a part of a program of instrumentation research oriented toward autommuled piine- tary experiments, supported by the National Aeronautics and Space Administration, research grant NGR-03.02)-004, Principal investigator, Joshua Lederberg; program director. Elliott C. Levinthal. P-?7; 1129 PRIVILEGED COMMUNICATION 2ECTION Il SUBSTITUTE THIS PAGE FOR DETAILED BUOCET PACE SUBSTITUTE PERIOD COVERED GRANT NUMBER FROM THROUGH DETAILED BUDGET FOR FIRST 12-MONTH PERIOD 1/1/74 12/31/74 1, PERSONNEL (List all personnel engaged on project) errant AMOUNT REQUESTED (Omit cents) NAME (Last, first, initial TITLE OF POSITION ”/ HRS. TOTAL Lederberg, J. Principal Investigator or 20 Program Director Kretchmer, N. Prof. of Peds. 10 pee Ge Nae ee Cann, H. Assoc.Prof. of Peds. 10 _ SCREENING FOR INBORN ERRORS Duffield,A. Research Associate | 100 |. ... OF METABOLISM. Pereira, W. Research Associate 100 pe Res Summons, R. Post.Doc. Fellow 100 Rindfleisch, T. Research Associate 20 Tucker, R. Computer Programmer 75 Veizades, N. Research Engineer 33 Steed, E. Research Engineer 33 Pearson, D. Elect. Tech. 40 Wegmann, A. Senior Res. Assist. 10 Open ~ Genetics Grad. Res. Assist. 50 Open - Genetics Grad. Res. Assist. 50 (continued) TOTAL ————____—-» |§ 157,208 2. CONSULTANT COSTS (Include Fees and Travel) § -_ 3. EQUIPMENT (Itemize) see attached list s 110,400* @& SUPPLIES See attached list s 16,200 TRAVEL oe. vomestic 2 East Coast (1000), 1 Mid-west (300), 1 West Coast § 1,500 (See Instructions) b. FOREIGN (200) $ 6. PATIENT COSTS (Separate Inpatient and Outpatient) s 7. ALTERATIONS AND RENOVATIONS Relocate equip., power, etc. for GC/MS Data System $s 1,800 8. OTHER EXPENSES (Itemize per instructions) Mini-computer maintenance $6,000 Freight on capital equipment $ 500 Office supplies, telephone, repro., postage, publication costs, etc. $2,600 9.100 s 3 9. Subtotal - Items Vithry 8 ee 18296, 208 10. TRAINEE EXPENSES (See Instructions) PREDOCTORAL No. Proposed $ FOR a. STIPENDS | POSTDOCTORAL No. Proposed $ OTHER (Specify) No. Proposed $ TRAINING DEPENDENCY ALLOWANCE $s GRANTS Z TOTAL STIPEND EXPENSES ———————» |$ b. TUITION ANO FEES s ONLY ¢. TRAINEE TRAVE'. (Describe) s iW Subtotal — Troinee Expenses $ 12. TOTAL DIRECT COST (Add Subtotais, Items 9 and 11, and enteron Page $296 208 Substitute Budget Page 5-72 wee fate FA P-73 GPO 930.75 continued 1. PERSONNEL TITLE Wyche, M. Lab. Tech. Boswell, M. Lab. Tech. Jamtgaard, R. Administrator Allen, M. Secretary - Genetics Murray, R. Secretary - Peds. Harlow, W. Machinist Open - Genetics Lab. Glasswasher + EQUIPMENT 4. Column Gas Chromatograph Mini-Computer System: PDP-11/45 (8k memory) 20,700 Memory (16k) 5,200 Floating Point Hdw. 5,200 Prog. Clock 600 Bootstrap Loader 600 Industry Tape 10,500 Disk System 11,700 Lab Interface Units 4,600 Interproc. Comm. 800 59 ,900 Amino Acid Analyzer Event Counter Digital Voltmeter Total Equipment SUPPLIES Chemicals,glassware, and lab apparatus GC supplies (columns, phases, etc.) Dry ice and liquid nitrogen Data recording media (GC/MS, Calcomp, etc.) Mini-computer supplies (start-up & continuing) Mass spectrometer repair parts & supplies Electronic parts & supplies Amino Acid Analyzer supplies Total Supplies % TIME 100 100 15 25 25 10 25 $14, 59, 32, 2, $110, $5, 1, i, 1, 2, l, 2,000 $16, 400% 900* 500* 700* 900* 400% 000 100 500 800 600 300 900 200 P-79 MS c TION If — PRIVILEGED COMMUNICATION SCREENING FOR INBORN METABOLIC ERRORS BUDGET ESTIMATES FOR ALL YEARS OF SUPPORT REQUESTED FROM PUBLIC HEALTH SERVICE DIRECT COSTS ONLY (Omit Cents) DESCRIPTION est PERIOD | ADDITIONAL YEARS SUPPORT REQUESTED (This application only) TAILED BUDGET) } 2ND YEAR 3RD YEAR 4TH YEAR 5TH YEAR 6TH YEAR 7TH YEAR PERSONNEL CosTs 157,208 | 168,338 } 215,892 230,758 | 246,632 CONSULTANT COSTS {Inciude fees, travel, etc.) EQUIPMENT 110,400*| 22,5004) 17,600% 5,300* 6, 000% SUPPLIES 16,200 16,900 20,200 21,400 22,800 DOMESTIC 1,500 1,800 1,900 1,900 2,000 TRAVEL FOREIGN PATIENT COSTS ALTERATIONS AND 1,800 RENOVATIONS OTHER EXPENSES 9,100 11,500 12,000 12,700 13,500 TOTAL DIRECT COSTS 296 ,208 221,038 | 267,592 | 272 5058 290,932 TOTAL FOR ENTIRE PROPOSED PROJECT PERIOD (Enter on Page 1, ten 4) ——$__—» $ 1,347,828 REMARKS: Justify all costs for the first year for which the need may not be obvious, For future years, justify equipment costs, as well as any significant increases in any other category. {f a recurring annual increase in personnel costs is requested, give percentage, (Use continuation page if needed.) Budget Explanation attached PHS-398 Ray, 2-70 Po &5 BUDGET EXPLANATION The early phases of the present study have been funded by NASA but that agency is rapidly severing its involvement in biological projects. It has already severely reduced its fiscal Support of our program in anticipation of total withdrawal within one or a few years. We have started to reduce laboratory statf in response to those cutbacks and will not be able to maintain this laboratory without new funding (like the present proposal). Since we cannot predict the level of funding we will receive for each of these proposals, we simply note the overlap and will negotiate a suitable joint effort for System development and operation, depending on funds available. The following is a detailed explanation of the budget for this subprogram. This budget covers only the indicated segment, with an indicated allocation for the tine of a number of participants who also appear in other segments. These allocations are integrated into the overall proposal budget summarized later (see Table of Contents for location). It should be noted that support for Professor Lederberg's time is shown at 20%. This includes not only his participation in this subproject but also his role in overall program direction as Principal Investigator. Salaries are increased at a rate of 6% per year to cover merit and cost of living increases. Staff benefits are applied based on the following University projections: 17%, 9/73-8/74; 18.3%, 9/74-8/75; 19.3%, 9/75-8/76; 20.3%, 9/16-8/773; 21.3%, 9/77-8/78; and 22.3%, 9/78-8/79. YEAR #1 (Metabolic Screening only) PERSONNEL: Support is requested for 10% of Professor Cann in Pediatrics (Professor Kretchmer will devote 10% of his effort without grant salary support) for clinical and diagnostic inputs to the program. Drs. Duffield, Periera, Summons, and Ms. Wyche are responsible for the derivatization and chemical analysis of urine samples including GC/MS operations. Support for an operator (Ms. Boswell) for the prescreening amino acid analyzec is included. Mr. Steed will maintain the glasswock aspects of the GC/MS system. Messers Rindfleisch and Tucker will be responsible for system design and software implementation respectively and Mr. Veizades, supported by fr. Pearson, will design the electronics. Ms. Wegmann operates the GC-High Resolution Mass Spectrometer in the bepartnent of Chemistry and will be available to run high resolution mass spectra on critically important Samples. Other part-time personnel (machinist, secretaries, and administrative Support) act in necessary supporting roles includiny assisting with liaison work with collaborating physicians at Stanford and elsewhere. -2- CAPITAL EQ“UIPMENT: We request a PDP-11/45 mini-computer systen aS an extension of the data system for our low resolution mass spectrometer. This system will initially have 24K words of memory and use a 1.2M word disk system for monitor, proyran, and spectral data storage. Removable data storage for raw data and archival spectrum recording is provided by the requested industry compatible magnetic tapes. The terminal, loader, clock, instrument interface hardware, etc. complete the system. Funds to maintain the computer system under manufacturer contract are included under "OTHER." A detailed computer equipment list is included in the "First 12 month budget detail." We also request a 4 colugsn gas chromatograph and an amino acid analyzer which will increase our urine prescreeniny capacity thereby alleviating the bottleneck at the mass spectrometer. The latter instrument is particularly pertinent to the larger scale prescreening of samples from the Kaiser Hospital, the Santa Clara Valley Medical Center, the Children's Hospital at Stanford, and the LA County Hospital. The amino acid analyzer will also allow reliable continued access to screening facilities for routine clinical studies during times when the GC/MS may be tied up for more sophisticated studies, or may be down for maintenance, repair, or re-engineering. The reliability offered by this back-up will help resolve the problem faced by any clinician when he attempts to investigate hew technigues while committed to offering the highest Standard of care. Once an analytical technigue is coupled to patient care, it cannot ethically be disengaged - a priority that pats great strains on any system also intended for further cesearch. In the present context, once a metabolite has been identified, it may be efficient to monitor its level, in repetitive sanples, by techniques like the amino acid analyzer. We have access to both low resolution and high resolution mass spectrometers (physically located in the Genetics and Chemistry Departments respectively) and do not request augmentation of this costly hardware for the present proposal. SUPPLIES: Supplies are requested for sample preparation and equipment operation. These include chemicals, glassware, laboratory apparatus, GC supplies, amino acid analyzer supplies, dry ice and liquid nitrogen, recorder paper for the GC/MS instruments and Calcomp plotters, and computer operation Supplies. In addition, we request supplies and parts for the maintenance (by our own hardware personnel) of the GC/MS systems and electronic eguipment as well as for the implementation of necessary data system interface hardware. TRAVEL: Travel costs are estimated at round trip expenses for two protessional meetings on the east coast and 1 trip each to the mid-west and west coast. -3- ALTERATIONS AND REMOVATIONS: We will need to aodify the electrical service and relocate some equipment in order to install the new data system and gas chromatograph and to allow for future equipment we are requesting. This estimate is based on a similar modification just made in another laboratory. OTHER EXPENSES: Office supplies, telephone service, postage, publication costs, etc. are requested based on operating experience in our laboratory. Computer aaintenance and equipment freight are estimated for the proposed data system augmentations. YEAR #2 PERSONNEL: Personnel are unchanged in year 2. CAPITAL EQUIPMENT: As the data system adaptation reaches completion after year 1, we request additional machine features to facilitate operational use simultaneously for GC prescreening and for MS analyses. These include a block of fast memory (1K words) to speed up the machine, an additional disk drive for larger volume on-line storage requirements and a high speed display device for system status and processing result feedback. Appropriate equipment maintenance increases are included for the expanded data systen. SUPPLIES, TRAVEL, AND OTHER: These items are increased by 6% to retlect inflation and the needs of increased throughput of urine screening. YEAR #3 PERSONNEL: We anticipate an increased volume of saaples to be processed by the third year as our baseline studies progress and more extensive clinical screening is undertaken. This implies the need for additional support both at the clinical and laboratory level. For this reason we have added 50% time of an Assistant Professor of Pediatrics and 100% time for a laboratory technician for sample preparation and equipment Operation to the budget. CAPITAL EQUIPMENT: An additional prescreening GC is requested to allow larger population analyses along with the appropriate computer interface equipsent. SUPPLIES, TRAVEL, AND OTHER: The budget for supplies and equipment maintenance are increased consistent with personnel and equipment additions. Also provision is made for inflation and longer daily operation. YEAR #4 PERSONNEL: Personnel remain unchanged in year 4. -4- CAPITAL EQUIPMENT: We request a system for the vacuum distillation of aqueous samples for specimen preparation and derivatization as well as miscellaneous test equipment which will require replacing. SUPPLIES, TRAVEL, AND OTHER: Increases are requested to cover expected inflation. YEAR #5 CAPITAL EQUIPMENT: A small provision is made for replacement test equipment. SUPPLIES, TRAVEL, AND OTHER: Increases Similar to year 4 are requested. SECTION III Maternal Blood Stream — Another Source of Fetal Tissue for Pre-Natal Diagnosis of Genetic Disorders Drs. Herzenberg and Cann P- 2S The #aternal Bloodstream - Another Source of Fetal Tissue for Prenatal Diagnosis of Genetic Disorders De. Le. A. Herzenberg, Principal Investigator Dr. H. Cann, Associate Investigator A. INTRODUCTION A.1 Objectives We propose to study methods of labelling, detecting and isolating fetal lymphocytes which pass into the maternal circulation during gestation. Further data on the time, extent and conditions of occurrence of these cells during normal human pregnancy could lead to non-invasive methods for prenatal diagnosis of genetic disease. A.2 Background and Rationale Fetal lymphocytes have been reported to be present in the mother's blood in the majority of pregnancies; they may be present in the maternal circulation in all pregnancies. The frequencies of fetal cells found, that is the percentage of tetal lymphocytes in the pregnant woman's blood streas, have ranged from 0.05% to as high as 1.5% in two studies. in one, by Walknowska, Conte and Grumbach (1), the presence of 5 small acrocentric chromosoaes was sought in lyaphocytes obtained fron 10 milliliters of maternal blood, stimulated by phytohemagglutinin (PHA) and examined at metaphase. Frop one to three metaphases out of many hundreds, suggested a male fetus in 21 pregnancies. This diagnosis was checked at birth and 19 males were correctly predicted. Three of these mothers were studied at 20 weeks or less of gestation. In the second study (2), quinacrine fluorescent-staining Y bodies were looked for in interphase lymphocytes obtained from blood (20 milliliter specimens) of women in the second and third trimesters of pregnancy. Here eleven male infants were predicted and seven weie verified at birth. Since the method used by Walknowska et al. is relatively insensitive, we suggest that fetal lymphocytes are present in tne maternal bloodstream in frequencies of about 0.1% or more in all pregnancies at least from 14 weeks of gestation on. Fetal lymphocytes possess HL-A antigens perhaps as early as 12 weeks of gestation (3). The enormous amount of polymorphisa noted in Caucasian populations (4) is such that most individuals will be heterozygous at one or both HL-A sub-loci and the likelihood of mates differing at one or both sub-loci is high. These facts suggest that fetal lymphocytes may be separated fror the maternal bloodstream by taking advantage of the HI-A differences existing between parents. A cytotoxic antibody directed toward the maternal HL-A antigen which the fetal lymphocytes do not carry, will, in the presence of complement, lyse the maternal leukocytes sparing those of the fetus. Or, -2- indirect labeling of the fetal lymphocytes coated with antibodies to the paternal HL-A type with fluorescein conjugated anti-human globulin will render these fluorescent in a sea of non-fluorescing maternal cells. The fluorescent cell-separator perfected by engineers working under the direction of one of us (Herzenberg) can separate flnoorescent from non-fluorescent cells with high efficiency, viability and yield (Figure 1; also reference 5). It can detect fFluorescent-labeled cells present in frequencies lower than 10**(-5) and can separate and thus enrich by a factor of 10**3. In model experiments with human erythrocytes, Rh+t cells stained with anti-D and fluorescein-conjugated anti-human immunoglobulins were detected about one time in two at a frequency of one positive cell per 10**5 kh negative cells. Post-partom blood samples of Rh- aothers who just gave birth to an Rht child were found to hava from 1/163,000 to 1/5,000 kKh+ cells indicating fetal blood leakage, probably at birth, of from 20 to 680 microliters (6). Thus, it should be quite feasible to enrich fetal lymphocytes fro# the reported frequencies of 10**(-3) to close to purity after the fetal cells are stained by inmunofluorescence for paternal HL-A antigens. We recognize the potential of the cell separator to separate fetal red blood cells from the maternal bloodstream on the basis of heritable erythrocyte antigens. Such nethodology could lead to the developuent of prenatal diagnostic tests for sickle cell anemia, possibly thalassemia, and other hereditary disorders of the erythrocyte. We have decided, however, to emphasize isolation of fetal lyaphocytes in this research project because of the reports (1,2) suggesting that these cells are detectably present in the maternal bloodstream in most, if not all, pregnancies. The literature suggests that fetal red blood cells are detectable in the maternal circulation in 5-10% of pregnancies at 20 weeks of gestation or less (7,8). Whether the application of the cell separator to detection and separation of fetal red blood cells will improve these figures, is a question we have been discussing. We believe that more data on the frequency of mothers showing fetal red blood cells in their circulations and the frequency of the cells themselves at various times in pregnancy are needed, and the cell separator could be applied to this research question. At present, our estimates of the amount of fetal red blood cells we can isolate, based on the performance of the cell separator and on the amount of fetal blood reported in the maternal circulation, indicate that our yield will be too low (perhaps 160 times too low) to permit diagnostic studies based on incorporation of radioactive label into globin synthesized by fetal cells (9,10). For these reasons too, we are persuaded to work with fetal lymphocytes. However, since we developed and used the methodology to detect gh- fetal cells in post partum Rh+ maternal blood specimens (6), we are planning preliminary experiments designed to answer questions as to amount, frequency, and requirements, raised above. We are collaborating with Dr. Herbert Schwartz, Professor of Pediatrics and Chief of the Pediatric Hematology Service at Stanford, in this preliminary -3- work. Dre. Schwartz will review the morphology of and determine whether globin synthesis can ba measured in cells separated frop the maternal circulation. Our preliminary work will be aimed at ways to increase yield and/or decrease amount of fetal blood required for prenatal detection tests. At this time, we are not asking for budgetary support in this area, although we may, should preliainary work show promise. The remainder of this proposal will deal only with the separation of fetal lymphocytes from the maternal circulatic.. B. SPECIFIC AIMS We propose to use heritable surface antigens as a basis for separating fetal lymphocytes from the maternal bloodstrean during gestation. These separated cells will be used to diagnose various genetic abnormalities in the fetus. C. METHODS OF PROCEDURE C.1 Immunoselection of Lymphocytes (a) POSITIVE LTMMUNOSELECTION WITH THE FLUORESCENT CELL SEPARATOR. Specific fluorescent staining for a variety of HL-A antigens will be developed. In most cases, we will use the indirect procedure, i.e. anti HL-A whole serum followed, after washing, by fluorescein or rhojamine conjugated anti-human immunoglobin (F or & anti-Ig). In pilot studies for this project, we have already obtained staining sufficiently bright to achieve excellent cell separation with a polyspecific and two specific HL-A antisera. In order to decrease non-specific staining (of lymphocytes lacking the antigens to which the antibodies are directed) we have had to remove aggregated proteins of antiserun by centrifugation at 100,000 gy for 2-3 hours. Lymphocytes are prepared by ficoll-Isopaque isopycnic centrifugation (11) followed by filtration through nylon wool to remove monocytes and many B-lymphocytes. Since these latter have surface immunoglobins, they stain with the fluorescent reagents, although much more dimly than the HL-A stained cells. with these virtually pure lymphocyte population, of which greater than 90% are thymus-derived (T) lymphocytes, we have obtained bright speckled fluorescence by staining at 0 degrees C. When we have passed positive and negative (stained for a particular HL-A type) cells separately through the c2l1l separator, the fluorescence intensity distribution has shown two clear populations with very little overlap (Figure 2). Thus, good separation can be obtained. Initially, reconstruction experiments will be performed using various mixtures of cells positive and negative tor an HL-A antigen or antigens. In addition to numerical evidence (from the fluorescence intensity distribution) for successful separation, ve will use male and female cells so that sex chromatin, karyotyping and quinacrine staining for ¥ bodies (2) can be useu as confirmatory markers. -- In isolations from maternal blood, antibodies directed to paternal and not maternal antigens will be used followed by F anti-Ig. (b) NEGATIVE IMMUNOSELECTION WITH COMPLEMENT MEDIATED LYSI3. AS an alternate procedure to or a step preliminary to fluorescent cell separation, (maternal) cells with particular HL-a antigens will be killed by incubation with antiserum and complement. Antisera directed against maternal antigens not shared by the father will be used when possible. With separate aliquots of lymphocytes, antisera negative for one or the other set of paternal antigens will be used. C.2 Fetal Lymphocyte Antigens as Alternates to HL-A for Iemunoselection with the Fluorescent Cell Separator Several fetal antigens have been described (12) but, so far, none of lymphocytes. [f an antigen could be found to be present on early fetal lymphocytes but absent, or in greatly diminished amount, on adult lymphocytes, it would be considerably more convenient to use than HL-~A antigens. There would be no problem of typing the father and having to select for each of the paternal haplotypes. We will attempt to find a fetal antigen by direct immunization of other species with lymphocytes from aborted fetuses. Any antisera obtained will be absorbed with adult cells, including lymphocytes, to rendar it specific. Of course, we cannot predict in advance if this will succeed. Another means to look for fetal lymphocyte antigens is to use antisera raised against leukemias. Tusors often have fetal or embryonic antigens; whether leukemias do is unknown. We will test for thea with fetal and normal adult lymphocytes. C.3 Mitogens Preferentially Stimulating Fetal Cells. It is possible that some mitogens will be relatively more stimulatory to fetal than adult lymphocytes. If so, this would decrease the importance of maternal contamination of separated fetal cells. We will screen all the available lectins including PHA, PWH (pokeweed mitogen), concanavalin A, wheat germ agglutinin, etc. for mitogenic activity with cells from aborted fetuses of several ages and cells from adults. One precedent for differential mitogenesis is PHA which stimulates fetal mouse thymocytes 10 times more than adult mouse lymphocytes (13). Evidence of differential response to PHA and PWM by lymphoid cells of huaan newborns (cord blood) and adults suggests that this approach may he feasible (14). C.4 Lymphocyte Culture and Karyotyping. Separated lymphocytes will be grown in micro-culture plates with phytohemagglutinin (PHA) for 48 hours, colcemid added and chromosome spreads made 16-20 hours later. In preliminary experiments we have obtained high percentages of metaphases with ~5— 290,000 cells cultured in 250 microliters of medium. frior staining with anti HL-A does not seriously decrease (less than 2-fold) the mitotic index. We have been Successful in obtaining high quality chromosome spreads by standard techniques with these small numbers of cells in micro-cultures. We believe we could culture (in 10 microliters medium) and spread as few as 1,000 lymphocytes. However, that is yet to be tried. To check our ability to S@parate, grow and karyotype fetal lymphocytes, we will predict the sex of the fetuses and compare with the sex observed at birth. C.5 Cell Separations Fluorescent cells will be separated in the fluorescent cell sorter we have built (see Figure 1). Cells suspended in mediun pass in single file in a streaa past 2 laser light sources and 2 photodetectors. One laser and detector observes cell fluorescence and the other light scatter which is a measure of cell volume. When a cell with the desired fluorescence level and of the right volume is detected, a charge is applied to the stream with a delay of about 100 microseconds. The stream is broken into 40,000 uniform droplets per second downstrean of the observing point. Charges are applied when droplets containing the desired cells are just being formed so they become charged. fhey are then deflected by constantly charged deflecting plates into collecting tubes. The nugber of cells processed per unit time is generally inversely related to the enrichment obtained. With a frequency of desired cells of 10*#(-3), 10**4 cells per second can be processed to obtain about 100% purity of wanted cells. A yield of at least 20,000 desired cells per hour should be obtained. This is certainly enough to culture and karyotype. If the frequency of fetal lymphocytes in the maternal bloodstream is 10-fold lower, Zee. 10**(-4), then only 2,000 cells per hour would be obtained. This is probably a marginal number for culture. Furthermore, there would probably be prohibitable losses with so tew cells. However, there is no need to seek high purity. When lookiny for male karyotypes, or later for gtoss chromosome abnormalities, only a few successtully spread cells of the selected type are needed. If we predict 5% purity, we need to screen about 100 cells to be quite certain of seeing a few wetaphases from the selected cells. Thus, preliminary analysis of maternal Sapples will be used to determine the amount of cells to be put through the cell sorter. With the 10¥**(-4) frequency, which is at least 10 times lower than the published estimates, 2 x 10**7 lymphocytes from maternal blood will be processed per hour. This approximately represents the number in 10 milliliters of blood. For a routine, non-objectionable procedure, we feel this (or perhaps 20 milliters) is about the limiting amount of blood which could be used. Thus, for screening purposes, the minimum fetal lymphocyte frequency is probably Slightly less than 10**(-4). We will ~6§- establish what the real frequencies are in a reaSonable number of pregnant women at various stages of gestation. If the published values are correct, we will not be limited by fetal cell frequency. (For high risk pregnancies, it may be that even with a frequency of less than 10**(-4), it would be preferable to do the antenatal cytogenetic analysis with 50-100 milliliters of blood rather than by amniocentesis) . C.6 Strategy (and Summary) (a) HL-A specific staining by immunofluorescence will be developed. This is presently in progress. (b) Preliminary enrichment by cytoxicity with Hi-A antisera and complement will be assessed. (Cc) Separation from artifically mixed populations will be carried out using sex markers to confirm, (d) Petal lymphocyte frequencies will be determined at various stages of gestation. (ec) Fetal lymphocyte antigens will be sought. (f) Selective fetal mitogens will be sought. (g) If the frequency in late first trimester or early second trimester pregnancies is about 10** (-4), separations, culture and karyotyping will be done. Accuracy of prenatal sex prediction will be used to confirm separations. Time Table: Development of the method with artiticial mixtures and maternal blood will take between one and two years. Assessment otf frequency of fetal lymphocytes in maternal blood as a function of gestational time will take about six months. Subsequent steps involve Jeveloping enough reagents for a Significant trial at detecting chromosome abnormalities using high risk pregnancies. This stage will also require setting up a cooperative effort involving patients at Stanford and other nearby institutions (e.g. the University of California Medical Center, San Francisco; Dr. C.J. Epstein). This will involve comparing results from the cell separator technique with those from amniocentesis. We anticipate applying for supplementary funds after 3 years if this pilot project is successful. D. SIGNIFICANCE A routine method of antenatal diagnosis using 10-20 milliliters of maternal blood would permit mass screening tor genetic disorders, especially those involving chromosomal aberrations. The methods we propose could result in avoiding the use of amniocentesis even in high-risk pregnancies (amniocentesis would be difricult to apply for mass screening). -7- The cost of separation ani karyotyping could be brought into the range of a few dollars per test with the development of semi-automated micro-cultures and computer assisted karyotyping (Dr. C.J. Epstein, personal communication). The cost of the 0.5-1% of congenital chromosome abnormalities among births is fur higher than the cost of such screening. It is impossible to assign a uniformly acceptable cost in total human terms, but the actual costs to the family and/or society must be in excess of 100,006 dollars for each infant with Down's syndrome born. Screening to detect an extra chromosome 21 would yield 1 case in 600 at a cost of probably from 1,000-10,000 dollars. Thus, the cost would likely be justified even for only this one condition. Further work could permit biochemical and particularly cytochemical tests (15) for inherited disease to be performed on the separated fetal lymphocytes either before or after culture. We wish to stress that the use of maternal blood rather than amniotic flaid samples provides a non-invasive method of antenatal diagnosis. This methodology can eliminate the risks to fetus and smother which are encountered in amniocentesis. Py: -3- REFERENCES 1. Walknowska, J., Conte, F.A., Grumbach, M.M., Lancet 1, 1119, 1969. 2. Schroeder, J., De la Chapelle, A., Blood 39, 153, 1972. 3. Payne, R., Blood 19, 411, 1962. 4%. Bach, F.H., Science 168, 1170, 1970. 5. Hulett, H.R. et al. Science 166, 747, 1969. 6. Jan, W., Herzenberg, L.A. in preparation. 7. Cohen, F. et al. Blood 23, 621, 1964 8. Clayton, E.M. Jr., Feldhaus, W.D., Whitacre, F.E., Obstet. Gynecol. 23, 915, 1964. 9. Hollenberg, M.D., Kaback, M.M., Kazazian, H.H. Jr., Science 174, 698, 1971. 10. Kan, Y.W. et al., New England J. med. 287, 1, 1972. 11. Borjue, A., Scandinav. J. Clin. Lab. Invest. 21, Suppl. 97, 1968. 12. Anderson, N.G., Coggin, J.li. Jr., Cole, E., Holleman, J.W. (eds.) EMBRYONIC AND FETAL ANTIGENS IN CANCER, Vol. 2, Oak Ridge National Lab., Tennessee, 1972. USAEC Report CONF~720208. 13. Mosier, D.E., Nature New Riology 242, 184, 1973. 14. Montgomery, J.R., Mason, BE.O., South, M.A., Pediat. Res. 7, 365, 1973. 15. Galjaacd, H, Jahodova, M., Niermijer, M.F., Brux. Med. 52, 109, 1972. P- 73 -93- TITLES OF FIGURES Figure 1. Diagram of fluorescent cell separator. Figure 2. Demonstration of almost complete separation of HL-A stained and unstained cells. te pe. Sb -0/ PRESSURIZEO CELL RESERVOIR ¥ PRESSURIZED =— OUTER FLOW "oT RESERVOIR ULTRASONIC TRANSOUCER LASER For nat CA) LASER FOR eg URGING PULSE SIGNAL HEI EXCITATION NN CHT ~\ NN NX, }— _> — HELIUM NEON FLUORESCENCE § HELIUM NEON © LASER 0 PHOTODETECTOR § SCATTERED LIGHT eo DETECTOR 0 ° o ° re) CELL SORTER 3 9 SIMPLIFIED BLOCK DIAGRAM -2«v {| § +2KV o> % | e a FLUORESCENT 3 CELLS NON- FLUORESCENT é CELLS o | ; 9 EMPTY DROPLETS s COINCIDENCES sar Ht: DEBRIS AND UNWANTED CELLS Figure ¢ CELLS OF RELATIVE NO, O48-3/14/773 Frequency Distribution 4 HL-A2 positive "T™" Lymphocytes j on Weak anti HL-A2 in Normal Human Serum iP | Strong anti HL-A2 ho 920 O40 28a ~ 128 CHANNEL NUMBER Fluorescence Intensity 160 229 245 Zen FiGure 2 P= Vb PRIVILEGED COMMUNICATION SECTION II SUBSTITUTE THIS PAGE FOR DETAILED BUDGET PAGE GRANT NUMBER SUBSTITUTE PERIOD COVERED FROM THROUGH DETAILED BUDGET FOR FIRST 12-MONTH PERIOD 1/1/74 12/31/74 1, PERSONNEL (List all personnel engaged on project) EPtoee AMOUNT REQUESTED (Omit cents) NAME (Last, first, initial) TITLE OF POSITION ”%/HRS. . . : : | TOTAL Principal Investigator or bi Sp be Herzenberg, Leonard Program Director 10% OR Ne oe Cann, H. Assoc. Prof. Peds. | 10% FETAL CELLS IN MATERNAL Open - Pediatrics Research Associate | 100% CIRCULATION Hulett, H. Research Associate | 20% Eee Waters, Ruth Research Assistant | 100% Sakaguchi, S. Research Assistant | 502% Open - Genetics Engineering Tech. 504% Meyering, P. Secretary 10% TOTAL mene | $54 782 2. CONSULTANT COSTS (Inctude Fees and Travel) $ 3. EQUIPMENT (Itemize) Carbon Dioxide Incubator $2,500 Laminar Flow Hood 1,000 Fluorescence Microscope 9,000 ; 12,500 * % SUPPLIES Tissue culture supplies, (filters, trays, etc.), dishes, media, pipets, sera, etc. $ 7,000 STAFF eo. vomestic 1 East Coast meeting s 500 TRAVEL (Se Instructions) b. FOREIGN $ 6. PATIENT COSTS (Separate Inpatient and Outpatient) $ 7. ALTERATIONS ANO RENOVATIONS $ 8. OTHER EXPENSES (Itemize per instructions) Office supplies, telephone, repro., postage, publication costs, etc. s 1,000 3. Subtotal - Items Vthry 8 > |$ 75,782 10. TRAINEE EXPENSES (See Instructions) PREDOCTORAL No. Proposed s FOR o. STIPENDS | POSTDOCTORAL No. Proposed || § OTHER (Specify) No. Proposed $s TRAINING DEPENDENCY ALLOWANCE $ GRANTS : TOTAL STIPEND EXPENSES ————————-» | $ b. TUITION AND FEES Ss ONLY €. TRAINEE TRAVEL (Describe) s WW Subtotal ~ Trainee Expenses $ . TOTAL DIRECT COST , , t —_—_—_-_—_—____> | 5 42, TOTAL O Cc (Add Subtotals, Items 9 and 11, and enter on Page 0 75 »782 Substitute Budget Poge 572 P-??7 SECTION It — PRIVILEGED COMMUNICATION FETAL CELLS IN MATERNAL CIRCULATION BUDGET ESTIMATES FOR ALL YEARS OF SUPPORT REQUESTED FROM PUBLIC HEALTH SERVICE . DIRECT COSTS ONLY (Omit Cents) DESCRIPTION iSAMeE AS pe. ADDITIONAL YEARS SUPPORT REQUESTED (This application only} TAILED BUDGET)! 2ND YEAR | 3RD YEAR 4TH YEAR 5TH YEAR 6TH YEAR 7TH YEAR PERSONNEL COSTS 54,782 | 58,659 | 62,704 | 67,021 | 71,629 CONSULTANT COSTS (include fees, travel, etc.) EQUIPMENT 12, 500* SUPPLIES 7,000 7,500 7,900 8,400 8,900 DOMESTIC 500 600 600 600 700 TRAVEL FOREIGN PATIENT COSTS ALTERATIONS AND RENOVATIONS OTHER EXPENSES 1,000 1,100 1,200 1,200 1,300 TOTAL DIRECT COSTS 75,782 67,859 72,404 77,221 82,529 TOTAL FOR ENTIRE PROPOSED PROJECT PERIOD (Enter on Page 1, item 4) —————_—» $ 375,795 REMARKS: Justify ail costs for the first year for which the need may not be obvious. For future years, justify equipment costs, as well as any significant increases in any other category. If a recurring annual increase in personnel costs is requested, give percentage, (Use continuation page if needed.) Budget explanation attached. PHS-398 > a OT, rt TG P-98 BUDGET EXPLANATION The budget covers 10% each of Professor Herzenberg's and Professoc Cann‘s time in relation to this subproject. A full time Research Associate is required to obtain, prepare, and fluorescent label the lymphocytes for separation and then to culture and perform cytogenetic studies on the separated cells. The part time (20%) Research Associate is to be responsible for the actual separations. Two Research Assistants, one part-time (50%) for cytogenetic studies and ongfull-time for serological studies, as well as a part-time (50%) cell separator operator are required to assist in these procedures. Support is also budgetted for 10% of Professor Herzenberg's secretary. Salaries are increased at a rate of 6% per year to cover merit and cost of living increases. Staff benefits are applied based on the following University projections: 17%,9/73-8/74; 18.3%, 9/7T4-8/75; 19.3%, 9/75-8/76; 20.3%, 9/76-8/77; 21.3%, 9/77-8/78; and 22.3%, 9/78-8/79. The laminar flow hood is essential for aseptic culturing and protection of personnel trom possible infectious agents in human blood. The carbon dioxide incubator is required for culturing lymphocytes. The fluorescent microscope will be used to detect interphase fluorescent Y¥ bodies and the Y¥ chromosome at metaphase. In the later stages of this study, the fluorescence microscope will be used to study the karyotypes of lymphocytes of fetuses with chromosomal aberrations. The budget for this project represents the MINIMUM ADDED COSTS to Dr. Herzenberg's and Dr. Cann's other programs {cell separator project, in particular) required for this work to be performed. P79 SECTION IV Polymorphic Genetic Markers in Amniotic Fluid Drs. Cann and Tsuboi P-/100