Micropuncture localization of kallikrein secretion in the rat nephron

Kidney International, Vol. 32 (1987), pp. 26—30

Micropuncture localization of kallikrein secretion in the rat nephron DEBBIE BEASLEY, NARENDRA B. OZA, and NORMAN G. LEVINSKY Renal Section, Evans Memorial Department of Clinical Research and Department of Medicine, University Hospital, Boston University Medical Center, Boston, Massachusetts, USA

Micropuncture localization of kallikrein secretion in the rat nephron. We have used free—flow micropuncture to study the tubular locus at which kallikrein enters the urine. Kallikrein was measured by a newly developed, very sensitive assay for kininogenase activity; active kallikrein was measured directly by this assay and total kallikrein after activation of inactive kallikrein. Kallikrein was readily detected in all of

Methods

Micropuncture experiments Experiments were performed on male Sprague—Dawley rats which were fed Purina rat chow (1% NaCl) and tap water ad 17, late distal tubular fluid—samples. In contrast, kallikrein was too low libitum until the morning of the experiment. They were anesto detect in 15 of 17 proximal or in 11 of 14 early distal tubular fluid samples. Calculations indicate that less than 10% of urinary kallikrein thetized with mactin (100 mg/kg) and prepared for snicropunccould have derived from filtration or from proximal secretion of ture as previously described [10]. The left ureter and bladder kallikrein. We conclude that urinary kallikrein enters the urine via were cannulated for collection of urine from the left and right secretion in the distal tubule. Filtration or proximal secretion of kidney, respectively. Tubular fluid collections were made at kallikrein does not contribute significantly to urinary kallikrein excre- least 30 minutes after surgery and only if proximal transit time tion. was less than 15 seconds and blood pressure was greater than 90 mm Hg.

Early proximal, late proximal, early distal and late distal segments were identified by i.v. injection of 0.05 ml FD&C Blue

Microdissection and histochemical observations indicate that dye. "Early" and "late" proximal tubules were identified as the first and last superficial loop of a given nephron. "Early" renal kallikrein is predominantly localized to the distal portions of the nephron, with especially high concentration in the and "late" distal refer respectively to the first and the last distal connecting tubule segments [1—7]. Little or no enzyme is segment in which dye appeared within a microscopic field. Each collection was made for 5 to 30 minutes, using an oil block to

present in the proximal tubule, loop of Henle, medullary permit complete collection of tubular fluid. Individual distal collecting ducts or glomeruli [3—6]. These data suggest that tubular fluid samples were assayed for kallikrein. For proximal kallikrein is secreted into the urine principally by the connect- samples, four to eight collections were pooled in order to ing tubule. In addition, a distal locus of secretion has been determine whether kallikrein was present in proximal tubular

demonstrated by a stop—flow study in the dog, which indicated fluid at low levels which otherwise might be below the limit of that kallikrein is added to distal tubular fluid [8]. However, detectability of the assay. Samples for active and total kallikrein these data do not preclude the possibility that urinary kallikrein measurements were collected from separate groups of rats (N = represents a combination of renal filtration of circulating tissue 17 and N = 10, respectively).

kallikrein and kallikrein secreted by the renal tubules. In fact studies by Fink and Schleuning [9] have suggested that renal clearance of circulating kallikrein may account for as much as 66% of urinary kallikrein.

Microkininogenase assay

A microanalytical modification of our standard assay for urinary active kallikrein [11] was developed (Oza et al, in

We have developed a specific kallikrein assay which is preparation). Briefly, in this assay tubular fluid was incubated with substrate and the kinins generated were analyzed by a

sensitive enough to detect kallikrein in tubular fluid samples obtained by micropuncture. Using free—flow micropuncture methods we have demonstrated directly that kallikrein is secreted into distal tubular urine. We have also found that little or no urinary kallikrein is derived from other sources, including filtration of circulating glandular kallikrein.

sensitive radioimmunoassay (RIA). Tubular fluid, in nanoliter volumes, was delivered directly from the collection pipette into

25 pi of buffer and stored at —70°C until analyzed. The substrate was pure bovine, low molecular—weight kininogen (Seikagaku Kogyo Company Limited, Japan). The tubular fluid

in 25 .d buffer and 10 g kininogen were incubated for five hours at 37°C. The reactions were terminated by heating at 100°C for five minutes. The kinins were extracted with ethanol.

Received for publication August 20, 1986 and in revised form November 25, 1986

The ethanol was evaporated and the kinins redissolved in buffer. Duplicate aliquots were analyzed by our kinin RIA,

© 1987 by the International Society of Nephrology 26

27

Kallikrein secretion in rat nephron

modified to enhance sensitivity. Modifications were: 1.) a highly sensitive bradykinin antiserum; and 2.) separation of bound and free antigen was accomplished with the use of particle—bound, goat—anti-rabbit immunoglobulins. Using these modifications, the sensitivity of the kinin RIA was increased about 25-fold and

70

60

the standard curve was now obtainable between 1 and 100 pg/tube. Several substrate blanks and internal standards (dilutions of a well characterized rat urine pool) were always run with each assay. The microkininogenase assay had intra-assay variability (C.V.) of 15.8% (N = 7 replicate analyses) and inter-assay C.V. of 18.6% (N = 10 replicate assays). Recovery of standard bradykinin was 108% 4% mean SEM (N = 9). In preliminary experiments, tubular fluid samples were assayed without incubation with substrate; preformed kinin was not detectable.

50 •0 (3 '3)

40

.

Activation of inactive kallikrein in tubular fluid 30

High purity Triton X-100 (Pierce Chemical Company, Rockford, Illinois, USA) was used to activate inactive kallikrein in the tubular fluid samples, using the method reported by Erdos and Yamada [121 for activation of membrane bound kallikrein. Tubular fluid samples in buffer and 5 pA of diluted Triton X-l00 were mixed, so that the effective strength of the Triton was 0.1% vol/vol. The activated tubular fluids were then incubated with kininogen and kinin assayed as described in the previous section. Substrate blanks containing Triton were always used in these assays. Activation of kallikrein by tritonization was documented in two ways. First, we compared the activation of kallikrein in a well—characterized rat urine pool by this method and by trypsin. Before activation, kininogenase activity was 71 3 pg kinin/5 hr incubation/nl urine. It increased to 104 6 after tritonization and to 115 17 after trypsinization. In addition, we determined the ratio of active and total kallikrein in ureteral urine of 10 rats used for the present micropuncture studies, using the tritonization method. Active kallikrein was 67% 9% of total kallikrein excretion. We have previously reported [13] that active kallikrein is 66% to 70% of total kallikrein using trypsinization and determination of immunoreactive kallikrein. These data provide strong evidence that activation of kallikrein with Triton in the microassay is comparable to activation of urinary kallikrein by trypsinization, the standard method. Therefore, triton activated samples are reported as total kallikrein.

I

$ S

..

20

S

.S

10

S Late

Early Proximal

Late

Early Distal

Fig. 1. Rate of active kallikrein collection from nephron segments of

the rat. Symbols are: (S) kallikrein measurements greater than or equal to the lowest standard; (0) kallikrein measurements less than the lowest standard.

these pools represented 40 minutes of tubular fluid collection,

the calculated rate at which kallikrein appeared at the early proximal site was minimal compared to the corresponding rate at the late distal site. The remaining proximal pools which had no detectable kallikrein activity also represented long collection

All kallikrein measurements are reported as units of times, varying between 20 and 75 minutes. Similarly, the

kininogenase activity. One unit is defined as one pg of kinin majority of early distal samples had no detectable kallikrein generated during a five hour incubation with substrate. All activity. Although kallikrein was detectable in three of fourteen urinary kallikrein measurements were made on 1 to 2 nI of urine early distal samples, activity was low in two of the three using the same microkininogenase assay used for tubular fluid samples. Both active and total kallikrein was readily detected in samples. all 18 late distal samples tested. In Figure 3, the rate at which kallikrein was collected at the Results

late distal segment is compared to urinary kallikrein excretion from the micropuncture kidney. Urinary excretion has been divided by 30,000, as an estimate of the total excretion to be rates at which kallikrein was collected at each nephron segment "expected" in each nephron. This estimate is based on three are shown in Figures 1 and 2. Kallikrein measurements are asumptions: the rat kidney contains 30,000 nephrons [14], the represented as detectable (solid circle) if the kinin generating secretory rate in each nephron is equal, and all urinary kalactivity of the sample was greater than or equal to the activity likrein is secreted before the late distal micropuncture site. of our lowest standard. Kallikrein was detectable in only two Active kallikrein detected at the late distal puncture—site was early proximal pools out of 17 proximal pools tested. Since 48% 8% of the expected excretion rate per nephron. Total

Both active and total kallikrein were consistently found in late distal tubular—fluid, whereas little or no kallikrein was found at early proximal, late proximal or early distal sites. The

28

Beasley

el a!

30 -

.)

50

Q

..20-

..

,

Ip210-

I

. I : 20

10

P AO ——I

Early

Late

Proximal

— —I—

—

I Late

Early Distal

2. Rate of total kallikrein collection from various nephron segments. Symbols are as in Figure 1. Fig.

Late distal

Urine ÷ 30,000

Fig. 3. Comparison of kallikrein in late distal tubular—fluid and the calculated "expected" excretion rates per nephron. Symbols are: (•) active kallikrein; (D) total kallikrein.

kallikrein was 58% 17% of the expected excretion per nephron. Comparisons between active and total kallikrein in late distal

Discussion

Evidence for the renal origin of urinary kallikrein includes

demonstration of renal synthesis [15—17], localization of renal these measurements were made in different rats. By chance, kallikrein in distal tubular segments [1—6] and entry of kallikrein total late distal kallikrein—measurements were made in rats into the urine at the level of the distal nephron using stop—flow which had lower urinary kallikrein excretion—rates than rats in techniques [8]. Our data demonstrate directly and quantitawhich active kallikrein was measured in late distal fluid. Active tively that kallikrein enters the urine in large amounts between kallikrein excretion was 247, 418 and 530 x i03 units/mm in the first and last distal segments detectable on the surface of the three rats in which total kallikrein was measured in late distal rat kidney. This location fits with many studies on the localizafluid and 387, 507, 720, 772, 781, 888, 1320 and 2021 X i0 tion of kallikrein in distal tubular segments of the rat nephron. units/mm in the eight rats in which active late distal—kallikrein Using microdissection techniques, the highest content of renal kallikrein has been found in the connecting tubule with signifiwas measured. cant amounts also present in the distal convoluted tubule and Tubular fluid collection rates were 25 1 nI/mm (N = 8) for the cortical collecting duct [1—3]. Figueroa et al [71, using early proximal, 20 1 ni/mm (N 6) for late proximal, 7.4 immunoelectron microscopic techniques in the rat, found kal0.5 (N = 16) for early distal and 4.8 0.5 nI/mm (N = 17) for likrein was specific to connecting tubule cells which have a fluid are not meaningful in these studies (Fig. 1 vs. Fig. 2), since

late distal samples. SNGFR measured in a separate group of similar distribution along the distal nephron. rats (N = 8), was 32 1 nllmin determined by proximal tubule Since late distal tubule puncture—sites identified by dye collections and 29 2 ni/mm determined by distal tubule injection correspond to initial collecting tubule [18, 19], most collections. Active kallikrein excretion from the left micropunc- renal kallikrein should enter the urine prior to the late distal ture kidney (780 90 x io units/mm, including all rats studied, segment. Our studies confirm that a large amount of kallikrein N = 22) was not significantly different from that of the contra- does enter prior to the late distal site. However, when we lateral kidney (840 120 x i03 units/mm, N = 22), indicating compared the amount of kallikrein present in late distal tubular that manipulation of the kidney for micropuncture does not fluid to that expected to be secreted per nephron, assuming significantly affect kallikrein excretion.

equal release by 30,000 nephrons [14], we found that active and

29

Kallikrein secretion in rat nephron

total kallikrein detected in the late distal tubule were only about late proximal tubular fluid samples which could have escaped one—half of the expected release per nephron. There are a detection ranged from 0.4 to 1.0 units of kallikrein per minute of number of possible reasons for this difference. First, the late collection. This represents less than 2% to 7% of the kallikrein distal puncture site may not regularly be beyond the entire present in urine samples obtained from the same rat. Even the portion of the distal nephron which secretes kallikrein. Proud, two early proximal samples which demonstrated low but meaKnepper and Pisano [3] estimated that 33% of distal nephron surable activity would represent only 4% and 8%, respectively, kallikrein is located in the initial collecting tubule and cortical of the expected excretion per nephron. Thus our measurements collecting duct. Our data indicate that as much as one—half of indicate that filtration of circulating kallikrein contributes less the kallikrein in the urine may enter in cortical or medullary than a few percent of urinary kallikrein excretion.

segments beyond the last loop available for puncture on the surface of the kidney. Secondly, superficial nephrons may secrete less kallikrein than other nephrons. It is now well recognized that superficial nephrons differ in a number of functional and anatomical characteristics from midcortical and juxtamedullary nephrons. For example, in the midcortical and juxtamedullary zones several distal nephrons join, forming arcades. Although arcades are thought to be lined with initial collecting tubule epithelia in the rat [3], kallikrein secretion by the arcades may account for part of the difference between late distal and urinary kallikrein. Another possible reason for the

Acknowledgments We thank Dr. K. Shimamoto for bradykinin antiserum and Dr. H. Kato and Mr. N. Yamamura for a very generous supply of pure bovine LMW kininogen. The technical assistance of Debbie Sanders, Catherine Murphy and Elizabeth Osmanski and secretarial assistance of Betty Kinsella is very much appreciated. This research was supported in part by National Institutes of Health Grants HL 31417 and HL 18318.

Reprint requests to Dr. Debbie Beasley, Renal Unit E428, University Hospital, 75 E. Newton Street, Boston, Massachusetts 02118 USA

References

discrepancy between late distal kallikrein and urinary kallikrein

is that we underestimated the number of nephrons in the rat kidney. Estimates vary between 30,000 and 38,000, and evidence has been presented that 38,000 is a better estimate for the mature rat [20, 21]. By using a value of 30,000 for our calcula-

tions, we may have overestimated the expected release per nephron. Finally, it is possible that puncture of the tubular epithelium somehow reduces the rate of kallikrein release, although there is no specific reason to postulate such an effect. Kallikrein was detected in 3 of 16 early distal samples. The "early" distal tubule as defined by dye transit times is thought to correspond to epithelia of the distal convoluted tubule [18, 19]. The presence of kallikrein in a few early distal samples may indicate that some kallikrein is secreted by this or prior segments. It seems more likely that the few positive samples are

due to inaccuracies in identifying tubular segments by dye transit times. Fink and Schleuning [9] infused pig kallikrein into dogs and detected substantial pig kallikrein in the urine. From these data and with the assumption that infused pig kallikrein is handled like endogenous circulating tissue—kallikrein, they concluded

that filtered kallikrein may constitute up to 66% of urinary kallikrein in the dog. However, endogenous tissue kallikrein in plasma (believed to be derived from various organs including

pancreas, salivary glands, etc.) is largely bound to plasma inhibitors [22]. Inhibitor bound kallikrein is a large complex which probably is not excreted or metabolized by the kidney [23]. One objective of the present study was to determine what fraction of urinary kallikrein, if any, is derived from filtration of circulating tissue kallikrein. Kallikrein was not detectable in 15

1. Sirvsso JAy, SPICER SS, CHAO J, GRIMM L, MARGOLIUS HS: Kallikrein localization in rodent salivary glands and kidney with the immunoglobulin—enzyme bridge technique. J Histochem Cytochem 27:1567—1576, 1979

2. ORSTAVIK TB, INAGAMI T: Localization of kallikrein in the rat

kidney and its anatomical relationship to renin. J Histochem Cytochem 30:385—390, 1982 3. PROUD D, KNEPPER MA, PISANO JJ: Distribution of immunoreactive kallikrein along the rat nephron. Am J Physiol 244:F510—F5l5, 1983

4. TOMITA K, ENDOU H, SAKAI F: Localization of kallikrein—like activity along a single nephron in rabbits. Pflugers Arch 389:91—95, 1981

5. OMATA K, CARRETERO OA, ScIcLI AG, HACKSON BA: Localiza-

tion of active and inactive kallikrein (kininogenase activity) in the microdissected rabbit nephron. Kidney mt 22:602—607, 1982 6. MARCHETTI J, IMBERT—TEBOUL M, ALHENC—GELAS F, ALLEGRINI

J, MENARD J, MOREL F: Kallikrein along the rabbit microdissected nephron: A micromethod for its measurement. Effect of adrenalectomy and DOCA treatment. Pflugers Arch 401:27—33, 1984 7. FIGUEROA CD, CA0R5I I, SUBIABRE J, Vio CP: Immunoreactive

kallikrein localization in the rat kidney: An immuno-electron microscopic study. J Histochem Cytochem 32:117—121, 1984 8. Sclcu AG, CARRETERO OA, HAMPTON A, CORTES P, OzA NB:

Site of kininogenase secretion in the dog nephron. Am J Physiol 230:533—536, 1976

9. FINK E, SCHLEUNING M: Studies on the excretion of tissue kallikrein from blood into the urine. Hoppe—Seyler's Physiol Chem 363:S1331—S1339, 1982

10. DAVIDMAN M, LALONE RC, ALEXANDER EA, LEvINSKY NG:

Some micropuncture techniques in the rat. Am J Physiol

221:1100—1119, 1971 11. LIEBERTHAL W, OZA NB, BERNARD DB, LEVINSKY NG: The effect

of cations on the activity of human urinary kallikrein. J Biol Chem 257:10827—10829, 1982

of 17 pooled proximal fluid samples (4 to 8 nephrons per 12. ERDOS EG, YAMADA K: Prekallikrein, kallikrein and renin in membrane fractions of rat kidney. Cliii Exper Hyper A4:2083—2096, sample). To evaluate the possibility that non-detectable amounts of kallikrein could have contributed significantly to urinary kallikrein, we calculated the maximal amount of kallikrein which could have been present in each of the proximal pools and compared it to the rate of urinary kallikrein excretion in the corresponding rat. The lowest standard varied between

1982

13. WEINBERG MS. OZA NB, LEVINSKY NG: Components of the

kallikrein—kinin system in rat urine. Biochem Pharmacol 33: 1779—1782, 1984

14. SPERBER I: Studies on the mammalian kidney. Zool Bidr Uppsala 22:249—431, 1944

20 and 40 units of kallikrein among assays. Total collection time for each proximal pool varied between 20 and 75 minutes. By

15. NUSTAD K, VAAJE K, PIERCE JV: Synthesis of kallikreins by rat kidney slices. Br J Pharmacol 53:229—234, 1975

calculation, the maximum amount of kallikrein in the early or

Kallikrein—like activity in perfusates and urine of isolated rat

16. ROBLERO J, CROXATTO H, GARCIA R, CORTHORN J, DEVIT0 E:

30

Beasley et a!

kidneys. Am J Physiol 23 1:1383—1389, 1976 17. MILLER DH, CIIAO J, MARGOLIUS HS: Tissue kallikrein synthesis

rons. II. Filtration rates, intraluminal flow velocities and fractional water absorption. Pflugers Arch 308:260—276, 1969

and its modifications by testosterone or low dietary sodium.

21. BAaarr U, WALLIN JD, RECTOR FC, SELDIN DW: Influence of

BiochemJ2l8:37—43, 1984

volume expansion on single—nephron filtration rate and plasma flow in the rat. Am J Physiol 224:643—650, 1973 22. LAWTON WJ, PROUD D, FRECH ME, PIERCE JV, KEISER HR, PISANO JJ: Characterization and origin of immunoreactive glandular kallikrein in rat plasma. Biochem Pharmaco! 30:1731—1737, 1981 23, RABITO SF, SETO M, MAITRA SR, CARRETERO OA: Clearance and

18. BENGELE HH, EVAN A, MCNAMARA ER, ALEXANDER EA: Tubu-

lar sites of potassium regulation in the normal and uninephrectomized rat. Am J Physiol 234:F146—F153, 1978 19. WOODI-IALL PB, TISHER CC: Response of the distal tubule and

cortical collecting duct to vasopressin in the rat. J Gun Invest 52:3095—3108, 1973

20. BAINES AD, DEROUFFIGNAC C: Functional heterogeneity of neph-

metabolism of glandular kallikrein in the rat. Am J Physiol 248: E664—E668, 1985