Glycoconjugates as noninvasive probes of intrahepatic metabolism: II. Application to measurement of plasma α 1-acid glycoprotein turnover during inflammation

Glycoconjugates as Noninvasive Probes of Intrahepatic Metabolism: II. Application to Measurement of Plasma cy l-Acid Glycoprotein Turnover During Inflammation Marc K. Hellerstein

and Hamish

N. Munro

We describe a new method for measuring plasma glycoprotein turnover in vivo using carbohydrate-moiety labeling compared with precursor UDP-sugar monitoring. which allows estimation of the specific activity of the precursor at the biosynthetic site of the glycoprotein. This method has been applied to plasma alpha l-acid glycoprotein (AGPI kinetics, and has allowed direct quantitation of absolute synthesis rates of AGP in a non-steady state. Following turpentine-induced acute inflammation, AGP was found to undergo a maximum increase in plasma level of eight-fold with e 20- to 2S-fold induction in absolute synthesis rate peaking at 25 to SO h, and a concurrent 2.0 to 2.6-fold increase in fractional degredation rate. The changes in absolute synthesis rates were quite comparable both temporally and quantitatively to changes in hepetic AGP mRNA levels and gene transcription rates reported by others following turpentine inflammation, thus suggesting that AGP synthesis in vivo is predominantly regulated by the level of its mRNA. The carbohydrate moiety labeling method can be applied to other plasma glycoproteins to measure their kinetic parameters in the intact animal or human subject. Q1987 by Grwm & Sfraffon, Inc.

C synthesis in vivo using labeled amino acids as precurURRENT METHODOLOGIES

for measuring protein

sors suffer from a general problem, namely, absence of easy access to the specific activity of the immediate precursor for protein synthesis. Attempts to use other amino acid pools in place of tRNA-bound amino acids (for review see reference 1) suffer from the uncertain relationships between various free amino acid pools (eg, extracellular-free, intracellularfree, proteolytic-derived) and the true charging pool for tRNA-aminoacylation. These methodological problems provided the motivation to develop a novel approach for measuring plasma protein synthesis rates in vivo avoiding the uncertainties of amino acid pools. The new method exploits the fact that most plasma proteins are glycoproteins and that protein-bound galactose and secreted giucuronic acid (GlcUA) are derived from a common UDP-glucose pool (see Fig 1 in companion paper*). We have demonstrated that the specific activity of the immediate precursor of the galactose moiety of plasma glycoproteins, UDP-galactose, can be monitored in vivo using GlcUA conjugated to the xenobiotic acetaminophen and excreted in the urine.* The method is conceptually analogous to the bicarbonate-urea method of Swick as applied by McFarlane and by Reeve.‘*5 In both methods, a labeled precursor enters an intrahepatic metabolite (the guanidine moiety of arginine or the hexose moiety of UDPglucose), which partitions into a secreted metabolite (urea or the xenobiotic-glucuronide) and an incorporated metabolite (protein-bound arginine or galactose). The bicarbonate-urea method unfortunately has become increasingly problematic and cumbersome to apply, mainly due to the large size, complexity and slow turnover of whole body urea pools into which labeled urea passes after secretion from the hepatocyte.’ The carbohydrate moiety labeling method has the advantage that secreted acetaminophen-GlcUA has a plasma half-life of 21 minutes and has no tissue storage ~001s.~This permits studies of short-term changes, in contrast to the urea method for which a period of 48 to 60 hours is required for an isotopic plateau in urinary urea to be attained.6

Metebotism, Vd

36, No 10 (October). 1987:

pp 995-1000

Here we apply the carbohydrate moiety labeling method to LYl-acid glycoprotein (AGP) synthesis by the liver of the turpentine-inflamed intact rat. AGP, a plasma protein of high carbohydrate content,“’ belongs to the class known as “acute-phase reactants” whose plasma concentrations rise markedly in infectious and inflammatory conditions.“,” AGP responds with one of the most extensive changes in concentration and has been studied as a model of the induction of protein synthesis in vivo and in vitro by investigators using the production of a sterile turpentine abscess to induce experimental inflammation. Because of its high galactose content and facile induction in the rat by means of turpentine inflammation, AGP is a suitable protein for testing the new technique. Moreover, recent studies of hepatic AGP mRNA levels utilizing specific cDNA probes12-‘6 or cell-free translation systems”-*’ have characterized the AGP mRNA induction kinetics following turpentine inflammation and require to be correlated with in vivo rates of AGP synthesis; the relationship between AGP synthesis and previously established hepatic mRNA levels or AGP gene transcription rates after turpentine inflammation’3-‘7 was, therefore, of interest. We report here the time course and quantitative aspects of changes in AGP turnover following turpentine inflammation in the rat. MATERIALS AND METHODS

Infusion of “C-Glucose and Induction of Turpentine Inflammation

Male Sprague-Dawley rates had intrajugular silastic catheters placed under ether anesthesia as described in detail previously.2J8 While still anesthetized, turpentine (0.5 mL/lOO g BW) was given muscularly. Then an intravenous infusion with .45% saline was From the USDA Human Nutrition Research Center on Aging at Tufts, Boston. Address Reprint requests to: M.K. Hellerstein. MD, PhD, USDA Human Nutrition Research Center on Aging at Tufts, 711 Washington St, Boston, MA 0211 I. Q1987 by Grune & Stratton, Inc. 0026-0495/87/3610-0016$03.00f0

995

HELLERSTEINAND MUNRO

996

commenced and continued for the duration of the experiment. The animals had access to food until 8 hours before tracer administration. Acetaminophen and tracer were administered at intervals up to 80 hours after turpentine injection. Acetaminophen was infused at a rate of 20 to 30 mg/kg/h; [U-‘4C]-glucose as a primed infusion with a bolus of 17 PCi, then infusion of 10 &i/h. Urine was collected every 60 to 120 minutes. Plasma samples were obtained via the indwelling cannula, the total value not exceeding 2 mL. The animals were sacrificed after 8 to 12 hours of infusion after anesthetization with ether and decapitation; and blood was collected and plasma obtained. Isolation of AGP From Plasma and Measurement of Plasma Concentration Rat AGP was isolated from 1 to 4 ml of heparinized plasma using a two-step procedure.‘9 The AGP is > 98% pure by SDS-PAGE and appears uncontaminated by immunoelectrophoresis against anti-rat whole serum antibody. The yield of AGP from plasma by this method was used to represent its plasma concentration, since recovery of AGP standards added in known amounts to plasma subjected to this procedure has been shown to be z 90%.‘9 Isolation of Galactose From Plasma AGP and Measurement of Specijic Activity AGP-bound galactose was cleaved, and the galactose isolated, measured chemically and counted.* The released galactose is free of peptides or carbohydrate contaminants.* The fluorometric galactose dehydrogenase method used to assay recovered galactose” is sensitive to < 0.5 pg quantities and linear through 5 to 7 gg. Triplicate assays had a variance of * 3%, while variance of specific activity determinations was * 4%.

Isolation of Acetaminophen-GlcUA From Urine and Measurement of Specific Activity Acetaminophen-GlcUA was isolated from centrifuged urine by HPLC on a reverse phase C-18 radial compression column.2 Acetaminophen-GlcUA recovery was quantitated by the peak height with reference to authentic acetaminophen-GlcUA of known quantity and specific activity determined. Plateau specific activity’ in urinary acetaminophen-GlcUA was attained by 2 to 6 hours following the priming bolus of 100 min of [U-‘4C]-glucose given as a single injection followed by a constant infusion. A typical time-course of precursor GlcUA specific activity of one rat is shown in Fig 1. Such plots allow an average specific activity of the intracellular precursor to be calculated during each period of the infusion of labeled glucose. Calculation of AGP Kinetics To measure the effect of turpentine administration on AGP synthesis, we applied the standard two-pool precursor-product model as described by Waterlow et al.’ Label is introduced into pool A (represented here by hepatic UDP-glucose) and incorporated into pool B (plasma AGP-galactose). Under conditions in which pool A and protein B are in a steady state, the fractional turnover rate (K) of the protein can be obtained from changes over time (dt) in the specific activity (S,) of the protein compared with the specific activity (S,) of the precursor pool, using the following equation:’ dS,/dt

(1)

In the above model, pool B is assumed to be constant in size so that the fractional rates of its synthesis (KJ and breakdown (K.J are equal, and can therefore be expressed in a single value for fractional turnover rate (K). If such a steady state does not obtain, it is still possible to compute the fractional synthesis rate (K,) of the protein’ since determination of synthesis rate is used to provide turnover in equation 1 under steady state conditions. The equation for the fracrional synthesis rate (KJ of the protein thus becomes: dS,/dt

-

- K (S, - S,)

- K, (S, - S,)

(2)

This equation can be integrated on the assumption that any inconstancy of K “will not be a serious source of error unless the change in protein mass is very great.“’ Thus, equation 2 can be transformed to: Ss - S*(l - e-&l)

(3)

(4)

Ks -

Fig 1. Typical time-course of the specific activity of urinary GluCA during continuour infusion of [U-‘4C~lucoso into a rat during a 24-hour period following turpentine injection. A priming dose of [U-“Clgluoose representing 100 times the infusion lwel per min was injected as a bolus, followed by constant infusion at 10 PCi/ h. Urine was collected at intervals and the specific activity of the GlcUA moiety of acetaminophen-GleUA was determined. For each collection period, the height of the block provides the average specific activii, while the length of the block represents the time of collection. The mean throughout the entire 2Chour period was used to provide the average precursor specific activii of UDP-galactose.

t

in which Ss represents change in the specific activity of galactose in AGP after a given time (t), and S, - mean specific activity of the precursor (derived from the GlcUA moiety of acetaminophenGlcUA (Fig l), corrected to M, 180 using a factor of 1.82) during the time period (t) of labeled precursor infusion. The mean absolute synthesis rate, vB’,, (me/dl/h) is calculated from K, over the time-period multiplied by B, the mean plasma concentration of AGP over the same time period, namely

The mean absolute degradation rate v,,s was determined from the difference between the mean absolute synthesis rate vs’,, and the rate of change of plasma AGP concentration (AB) during time (t) as follows: v.4, - (v,,)

- (ABlt)

(6)

GLYCOCONJUGATES

AS NONINVASIVE

Table 1.

h+ctiit

Ratesof Synthesis and Breekdown

Rats

Basal (4)

COncMltr~tiOll

kng/dL~

No. of

of Alpha l-Acid Glyooprotein Following Turpentine Injection*

Rate of AGP

AGP Concentratims~

Time After Tupentim,

997

METABOLIC PROBES-II

Cha_ng&

Fractional Rate (d-‘h-l)

Absolute Rate (mg/rUh)

Bretl~tiwn

S@WiS

S~lbSiS

Bre4@own

Pre

Post

Mean

52

52

52

0

l.l(O.0481

l.l(O.046)

2.4

2.4

(AS/h)

(K,)

(K,)

(V,)

(V,,)

12 h (6-17)

2

62

80

71

1.6

1.2fO.050)

0.7fO.027)

3.6

2.0

14 h (9-19)

1

70

92

81

2.1

1.5fO.083)

0.9fO.037)

5.1

3.0

17 h (12-22)

2

82

137

110

5.5

2.6fO. 108)

1.4fO.058)

11.9

6.4

22 h (14-29)

2

89

241

165

10.1

3.6fO.150)

2.1(0.089)

24.8

14.7

26 h (20-32)

2

100

280

190

15.0

6.2f0.258)

4.4fO. 183)

49.1

34.7

31 h (25-37)

3

120

370

245

20.8

5.9iO.2461

3.9tO.16 1)

60.2

39.4

45 h (40-60)

4

320

405

365

8.5

3.1fO.129)

2.5(0.106)

47.1

38.8

6 1 h (52-70)

2

380

220

300

-8.9

0.7(0.029)

1.410.059)

8.8

17.7

8 1 h (72-90)

3

215

135

175

-4.4

0.7tO.029)

1.3fO.054)

5.1

9.5

*For formulas used in these calculations, see “Experimental Procedures.” TTime after turpentine injection (mean, start and finish). $Pre and post refer to the beginning and end of each labeling period. @ate of change in amount between beginning and end of labeling period. /IValues are given as rate/d: values in parentheses are given as rate/h.

The fractional degradation rate (K,,) was then determined directly using the derivation,’ I
(7)

RESULTS

We observed base-line concentrations of AGP averaging 52 mg/dL plasma in untreated rats (Table 1). Other workers using the rat have observed values from 6 to 9 mg/dL to 250 mg/dL in the basal state.‘2*‘3~22’29 AGP is variably antigenic*’ and the lower estimates have been observed using antibodydependent quantitation methods, such as the basal concentration of 6 mg/dL determined by monoclonal antibody.12 For this reason we chose an estimation based on physical rather than the antigenic properties of AGP.19 However, it should be noted that some workers using immunologic methods have reported similar or higher basal AGP concentrations compared to ours.24 When inflammation was induced by turpentine, plasma concentrations of AGP increased, attaining a peak of slightly greater than 400 mg/dL some 40 to 50 hours after turpentine, which represents an eightfold increase from the baseline concentration of 52 mg/dL (Table 1 and Fig 2). The AGP concentration still remained elevated at 280 to 300 mg/dL at 60 hours and then fell to 135 mg/dL by 90 hours after turpentine. The fractional increases in plasma AGP levels reported by others after turpentine inflammation have tended to be inversely related to the measured basal level and to range from a 5- to 15-fold increase, which is within the range that we report. Figure 3 shows the time course during induction and de-induction of the kinetic parameters K, and vaA (fractional and absolute synthesis rate respectively). It is evident that the K, rises and falls before the V,,, as anticipated. Thus at 45 hrs after turpentine, a K, of 3.1 still represents an absolute synthesis rate (va,,) of 47 mg/dL/h due to the elevated plasma AGP mass present (Table 1 and Fig 2). The 25-fold increase of total AGP synthesis from 2.4 mg/dL/h to 60 mg/dL/h should be emphasized. It has been shown26 that the

20% extravascular distribution of AGP remains unchanged after turpentine inflammation. Accordingly, at an average maximal synthesis rate of 40 mg/dlh. over 24 hours (Table l), 360 mg/kg/d of AGP is being synthesized By comparison, the basal rate of protein synthesis for all plasma proteins in the rat without inflammation has been estimated to be 3.1 to 4.3 g/kg/d.*’ Thus under stress, AGP synthesis can account for 10 percent of the export protein synthesized by the liver. The kinetics of the fractional breakdown parameter K,, (estimated from incorporation relative to change in plasma

400 r

0

12

24

30

42

I

00

72

04

Time After Tupen5ne 6-k.) Fig 2. Plasma AGP concentrations followi~ turPentineinduoed inflammation in the rat. Plasma AGP was detormbted as described in the text. Each point shown represents the mean * SEM from two to four animals.

HELLERSTEIN AND MUNRO

998

Haus

After

Turpentine

Administration

Time-course of changes in plasma AGP synthesis and Fig 3. AGP-mRNA abundance following turpentine-induced inflammation in the rat. K,, fractional synthesis rate (d-‘1, V,, absolute synthesis rate (mg/dl/h). mRNA level expressed relative to base1 uninduced abundance. The_mRNA values are adapted from the report of Ricca et $1.” K, end V,, represent mean values for two to four rats. When the rats were grouped on the basis of time after turpentine administration, the following statistical evaluation emerged (mean i SEMI: Group I (Basal) KS - 1.09 + 0.10, Group II (t-6-20h)~-1.80~0.43.GroupIIIft-20-4Oh~K,-6.6i 1.00,GroupIVk-40-Wh~K,-3.13i1.26.GroupVft-M)70 h) K, - 0.70 f 0.10. Group VI ft > 70 h) K, - 0.68 * 0.20. Comparison of Group I v Group II, P - .lO, Group I v Group Ill. P < .006. Group II Y Group Ill, P c .OO& Group II Y Group IV, P - .16, Group Ill v Group IV. P - .09. Group IV v Group V, P - .13. Group V v Group VI. P - 4%.

AGP concentration) can also be seen in Table 1. Other than two elevated values of 26 and 32 hours (concurrent with greater than twofold increases in plasma AGP concentrations over the experimental period), the K, values throughout most of the post-turpentine period appear to vary between about 1.0 and 2.5d-‘, although the estimated absolute catabolic rate V,, varies over a 20-fold range. Interanimal variability between paired animals catheterized consecutively and analyzed at the same time under the same conditions in general showed good reproducibility (not shown). DISCUSSION

We describe the application of a new method for measuring plasma glycoprotein turnover to the induction of AGP in the turpentine-treated rat. The model is described and validated elsewhere* and consists of labeling the hepatic UDP-sugar pools with infused glucose, and taking advantage of the precursor/product relationships that exist between UDP-glucose and both UDP-galactose and UDP-GlcUA. In consequence, the specific activity of the GlcUA can be used as an end-product to represent the specific activity of the intrahepatic UDP-glucose pool which is also precursor for UDP-galactose and thus of glycoprotein-galactose. Using the carbohydrate labeling method, the basal synthesis rate (IQ of AGP was found to average 1.1 d-’ (Table l), ie, a 7I/2 of 15 h. This estimate is close to but lower than those

of other workers using I ‘25-labeled AGP in rats in the steady state, namely r,,2 20 h,2823 hT4 and 32 h.26Also, concordant are reports of incorporation into AGP relative to albumin in vivo or in perfused liver,23~29~30 which have yielded K, values for AGP about 2.5 to 3.0 times that of albumin, whose T,,~ in the rat is about 2.5d. In conformity with the general finding in rats that their metabolic processes are about 5 times those of human,3’ the T,,~ in the AGP of the rat is about one fifth the T,,~ of AGP in the human, namely 3.5 to 5.0 d.32 The changes in fractional (sixfold) and absolute (25-fold) synthetic rates observed following turpentine inflammation (Table 1) are the first quantitative estimates of AGP synthesis rates in the nonsteady state following turpentine inflammation using a true precursor-product methodology. Most previous studies are based on total label incorporation into AGP or seromucoid?2,26,29.30,33 Schreiber et al26 reported a quantitative estimate at a single time-point and found a comparable 22-fold increase in absolute synthesis rate 24 hours after turpentine. However, they made no direct measurements of precursor specific activity, but relied on a number of indirect assumptions. It is of interest to compare our data for synthesis rate to previously reported changes in AGP mRNA levels.‘2-17Increase in mRNA abundance has been reported to vary between a 20-fold increase’-“” to 73-fold” or 90-fold.‘* One of these” showed that the increase was due to a comparable increase in AGP gene transcription rates in isolated hepatocyte nuclei. Thus, it has been assumed that AGP synthesis in vivo during the acute phase response is primarily regulated through augmented synthesis of its mRNA. Our data provide evidence for this. Finally, we have to consider the validity of the model used by us to measure synthesis and breakdown of a specific glycoprotein in a nonsteady state. Incorporation models such as ours can be shown to have an important mathematic characteristic, namely that they remain valid in the nonsteady state of protein mass (eg, during its growth or decay) so long as one measures specific activities of precursor and product rather than total radioactivity in the protein pool.’ However, the mathematical treatment of the non-steady state condition involves two linked differential equations with three parameters B (protein mass), K, (or KBA), and V,,. In the nonsteady state, B is by definition not constant; consequently, either K, or V,, must be constant for a mathematic solution to be possible. The assumption is often made that, over the period of study, K, is relatively constant, although this may be inaccurate during a rapid induction such as for AGP during inflammation. The error from this assumption is, as Waterlow et al’ put it “likely to be insignificant unless protein mass B changes dramatically,” but as can be seen (Table l), the periods between 26 hours and 31 hours exhibit precisely this combination of changes for the three parameters (major change in protein mass and changes in K and V,,). In essence, the turpentine-inflamed rat model is so responsive kinetically that it stresses the mathematic model for periods other than very short infusions during peak synthetic periods. The only mathematical consequence of this during rapid increases in protein pool B would be a possible overestimation of K,. Since the calculation of K, is directly dependent upon I&, it too would be overestimated

GLYCOCONJUGATES

AS NONINVASIVE

999

METABOLIC PROBES-II

in this setting. One can calculate that, for example, if K, at

26 h were 4.5 rather than 6.2, va, would still be 45.9 mg/dl/h. [Consistent with this estimate, the absolute synthesis rate determined at 46 h (Table 1 and Fig 3), which represented a period of relative stability in AGP mass, was in the range of 47 mg/dL/h] but VA, would be 25.1 mg/dL/h and Kd would be 2.46/d rather than 4.4/d. This is consistent with other reports of about twofold increased fractional catabolism using decay techniques following turpentine inflammation.23~2s The model we propose meets the following possible criti-

cisms: 1. That turnover of the carbohydrate (galactose) moiety of AGP be similar to that of the protein moiety. Both direct experimental evidence and theoretic considerations support the conclusion that they are identical in plasma glycoproteins. In particular, the Ashwell-Morel1 hypothesis34 regarding the role of the galactose receptor in plasma glycoprotein catabolism implies that galactose turnover determines turnover of the peptide-moiety, nor is there evidence for a change in the galactose content of AGP during the acute phase response.” 2. That the UDP-glucose pool of liver could be shown to be isotopically homogeneous with regard to galactosyl- and glucuronyl- conjugates, thus excluding isotopic compartmentalization between these secreted glycoconjugates. This was confirmed elsewhere,’ although UDP-glucose for glycogen appears to be in a separate compartment of the liver.‘* Prior

turpentine administration did not alter the finding of a common hepatic UDP-glucose for galactosyl- and glucuronyl-conjugates.’ Direct galactose entry by the galactose-1-P pathway could not be substantial because the two endproducts GlcUA and AGP-gal attain the same specific activity after prolonged “C-glucose infusion, which would not be possible if another source of labeled or unlabeled galactose entered the system.* This demonstrates that endogenous glycoprotein catabolism does not contribute significant galactose compared with the amount coming from glucose. 3. That the experimental conditions did not alter the acute-phase glycoprotein response to turpentine. Although acetaminophen is an analgesic and antipyretic drug, it possesses minimal anti-inflammatory action (Dinarello C, personal communication). The observation (Table 1) that we obtain increments in AGP similar to those seen by other authors not using acetaminophen would appear to rule out a suppressive action of acetaminophen given in moderate doses as a probe for hepatic GlcUA. In conclusion, the carbohydrate moiety labeling method can in theory by applied to any plasma glycoprotein containing sufficient galactose for measurement of specific activity for radioactive labels or enrichment using mass spectroscopy for stable isotopes. For example, galactose can be removed enzymatically from the apolipoprotein B of intact lowdensity lipoprotein particles (unpublished), making this a promising approach to the turnover of this important protein which has proved to be difficult to study by other methods.

REFERENCES

1. Waterlow JC, Garlick PJ, Millward DJ: Protein Turnover in Mammalian Tissues and in the Whole Body. Amsterdam, NorthHolland, 1978 2. Hellerstein MK, Greenblatt DJ, Munro HN: Glycoconjugates as noninvasive probes of intrahepatic metabolism: I. Kinetics of label incorporation with evidence of a common precursor UDP-glucose pool. Metabolism 36:988-994, 1987 3. McFarlane AS: Metabolism of plasma proteins, in Munro HN, Allison JB (eds): Mammalian Protein Metabolism, vol 1. Orlando, FL, Academic, 1964, pp 297-341 4. Reeve EB, Pearson JR, Martz DC: Plasma protein synthesis in the liver. Method for measurement of albumin formation in vivo. Science 139:914-916, 1963 5. Swick RW: Measurement of protein turnover in rat liver. J Biol Chem 228:751-764, 1958 6. Gersovitz M, Munro HN, Udall J, et al: Albumin synthesis in young and elderly subject using a new stable isotope methodology: Response to level of protein intake. Metabolism 29:1087-1096, 1980 7. Bennett M, Schmid K: Immunosuppression by human plasma alpha l-acid glycoprotein: Importance of the carbohydrate moiety. Proc Nat1 Acad Sci USA 77:6109-6113.1980 8. Costello M, Fiedel BA, Gewurz H: Inhibition of platelet aggregation by native and desialized alpha l-acid glycoprotein. Nature 28 1:667-668, 1979 9. Chiu KM, Martensen RF, Osmond AP, et al: Interactions of alpha l-acid glycoprotein with the immune system. Immunology 32:997-1005, 1977 10. Koj A: Acute phase reactants, in Allison TB (ed): Structure and Function of Plasma Proteins. New York, Plenum, 1973, pp 73-132 11. Gordon AH: The acute phase plasma proteins, in Bianchi R,

Marianna G, McFarlane AS (eds): Protein Turnover. Baltimore, MD, University Park, 1976, pp 381-394 12. Ricca GA, Hamilton RW, McLean IW, et al: Rat alpha l-acid glycoprotein mRNA. Cloning of double-stranded cDNA and kinetics of induction of mRNA levels following acute inflammation. J Biol Chem 256:10362-10368,198l 13. Diarra-Mehrpour M, Bourguignon J, Leroux-Nocollet I, et al: The effects of 1‘I-alpha ethynyl oestradiol and of acute inflammation on the plasma concentration of rat alpha l-acid glycoprotein and on the induction of its hepatic mRNA. Biochem J 225:681-687, 1985 14. Bauman H, Firestone GL, Burgess TL, et al: Dexamethasone regulation of alpha l-acid glycoprotein and other acute phase reactants in rat liver and hepatoma cells. J Biol Chem 258563-580, 1982 15. Kulkarni AB, Reinke R, Feigelson P: Acute phase mediators and glucocorticoids elevate alpha l-acid glycoprotein gene transcription. J Biol Chem 260~15386-15389, 1985 16. Schiaffonati L, Bardella L, Cairo G, et al: Mechanism of enhanced RNA synthesis in acute phase rat liver and its relationship to chromatin structure. Biochem J 219:165-171, 1984 17. Northemann W, Andus T, Gross U, et al: Messenger RNA activities of four acute phase proteins during inflammation. FEBS Lett 161:319-322, 1983 18. Hellerstein MK, Greenblatt DJ, Munro HN: Glycoconjugates as noninvasive probes of intrahepatic metabolism: Pathways of glucose entry into compartmentalized hepatic UDP-glucose pools during glycogen accumulation. Proc Nat1 Acad Sci USA 83:70447048.1986 19. Hellerstein MK, Sasak V, Ordovas J, et al: Isolation of alpha l-acid glycoprotein from human plasma using high performance liquid chromatography. Anal Biochem 146:366-371, 1985

1000

20. Winzler RJ, Dewar AW, Mehl JW, et al: Studies on the mucoproteins of human plasma. I. Determination and isolation. J Clin Invest 27:609,1948 21. Kornblatt MJ, Schacter H, Murray RK: Fluorometric galactose dehydrogenase determination. Biochem Biophys Res Commun 48:1489-l 490, 1972 22. Ritchie DG, Fuller GM: Hepatocyte stimulating factor: A monocyte-derived acute phase regulatory protein. Ann NY Acad Sci 390:490-502, 1983 23. Richmond JE: Studies on the metabolism of plasma glycoproteins. Biochemistry 2:676-683, 1963 24. Shibata K, Okubo H, Ishibashi H, et al: Rat alpha l-acid glycoprotein: Uptake by inflammatory and tumor tissues. Br J Exp Path01 59:601, 1978 25. Zeineh RA, Kukral JC: The turnover rate of orosomucoid in burned patients. J Trauma l&493,1970 26. Schreiber G, Howlett G, Nagashima M, et al: The acute phase response of plasma protein synthesis during experimental inflammation. J Biol Chem 257:10271-10277, 1982

HELLERSTEIN AND MUNRO

27. Jeffay H, Winzler RJ: The metabolism of serum proteins: I. The turnover rates of rat serum proteins. J Biol Chem 231:101, 1958 28. Kuranda MJ, Aronsen NA Jr: Tissue locations for the turnover of radioactively labeled rat orosomucoid in vivo. Arch Biochem Biophys 224:516-533,1983 29. Ashton FE, Jamieson JC, Friesen AD: Studies on the effect of inflammation on rat serum proteins. Can J Biochem 48:841-850, 1970 30. Sarcione EJ: Synthesis of alpha l-acid glycoprotein by the isolated perfused rat liver. Arch Biochem Biophys 100:516, 1963 3 1. Munro HN: Evolution of protein metabolism in mammals, in Munro HN (ed): Mammalian Protein Metabolism, vol 3. Orlando, FL, Academic, 1969, pp 133-182 32. Weisman S, Goldsmith B, Winzler R, et al: Turnover of plasma orosomucoid in man. J Lab Clin Med 57:7, 196 1 33. Bostrom H, Rosen L, Yamashira I: On the metabolism of the acid glycoprotein in plasma. J Biol Chem 230:381, 1958 34. Harford J, Shwell G: The hepatic receptor for asialoglycoproteins, in Horowitz MT, Pigman WW (eds): The Glycoconjugates. Orlando, FL, Academic, 1977, pp 27-56