• © 1997 Nature Publishing Group http://www.nature.com/naturebiotechnology
`
`RESEARCH
`
`Regulation of the biological activity of
`gluc·agon-like peptid~ 2 in vivo by
`dipeptidyl peptidase IV
`
`Daniel J. Drucker'·*, Qing ShF, Anna Crivici', Martin Sumner-Smith~, Wendy Tavares2
`Lorraine DeForest', Sari Cooper2, and·Patricla L. Brubaker'.1
`
`, Mary Hiii',
`
`Dtpartmmts of Medicini and Physiology, Banting and Best Diabetes CmlTe, The Toronto Hospita~ University of Toronto, Toronto, Ontario and
`Allelix Biopharmaceutical5 Intl., M'wis5auga, Ontario. *Corresponding author ( e·mail: cl.drucker@utoronto.ca).
`
`R.e<%ivcd 27 March 1997; accepted 13 May 1997
`
`Specle&-apeclflc dlffentncea In the enzymatic Inactivation of peptides la en Important consideration in
`the evaluation of therapeutic efficacy. We demonstrate that glucagon-llke peptide 2 {GLP-2), shown to be
`highly .lntestinotrophlc In mice, promotes an Increase in Intestinal villua height but haa no trophic effect
`on small bowel weight In rats. The reduced lntestlnotrophlc activity of GLP-2 In rats is attributable to
`Inactivation by the enzyme dlpeptidyl peptidase IV (DPP-IV). GLP-2(1-33) was degraded to GLP-2(3-33)
`following Incubation with human placental DPP-IV or.rat serum but not by serum from DPP-IY.deficlent
`rats. Administration of rat GLP-2 to OPP-IV-deficient rats waa aaaociated with markedly increased bioac(cid:173)
`tivlty of rat GLP-2 resulting in a significant Increase in small bowel weight. A synthetic GLP-2 analog,
`r[Gly'JGLP-2, with an alanine to glycine substitution at position 2, was resistant to cleavage by both DPP(cid:173)
`IV and rat serum In vttro. neatment of wild·lype rats with r[Gly'JGLP-2 produced a statistically significant
`Increase In small bowel mass. DPP·IV·medlated Inactivation of GLP-2 la a critical determinant of the
`growth factOl'-Dke properties of GLP-2.
`
`Keywords: protein modeling, protease inactivation, growth factor, intestine, therapeutic
`
`The differentiated cell types of the small bowel epithelium serve an
`important physiological function in the regulation of nutrient and
`electrolyte absorption. The rapid turnover and renewal of the
`intestinal mucosa! epithelium provides a unique model for the
`analysis of the molecular factors important for cell proliferation
`and apoptosis. A number of peptide growth factors have been
`shown to be trophic for the small bowel in vivo, and exert their
`effects in general via stimulation of crypt cell proliferation and
`inhibition of apoptosis'. Whereas many epithelial growth factors
`such as growth hormone, various interleukins, the insulin-like
`growth factors, and keratinocyte growth factor stimulate the pro(cid:173)
`liferation of a diverse number of epithelial cell types, peptide hor(cid:173)
`mones such as gastrin or neurotensin appear to exert their
`growth-promoting effects in a more tissue-restricted manner.
`The peptides coencoded with glucagon in the proglucagon
`gene have long been implicated in the control of intestinal growth
`and regeneration. The clinical reports of patients that presented
`glucagon-producing tumors and ·villus hyperplasia of the small
`bowel"' were followed by experimental observations linking
`intestinal growth with increased production and secretion of the
`proglucagon-derived peptides•. Furthermore, major small bowel
`resection in rats is associated with increased expression of the
`proglucagon gene in the intestinal remnant', and bowel resection
`in humans is associated with inaea~d release of enteroglucagon
`following a test meal'; providing further albeit indirect, evidence
`that links intestinal adaptation with enhanced production of the
`proglucagon-derived peptides (PGDP) in vivo. Nevertheless,
`despite numerous &tUd.ies demonstrating an associ11tion between
`proglucagon gene expression and intestinal growth, the identity of
`the PGDP with intestinotrophic properties remained unclear.
`Th elucidate the relationship between peptides encoded within
`
`NATURE BIOTECHNOLOGY VOLUME 15 JUl.Y 1997
`
`proglucagon and small bowel growth, we established a series of
`transplantable glucagonomas in nude mice. Elevation of the circu·
`lating PGDPs in tumor-bearing mice7 was consistently associated
`with the induction of crypt cell proliferation leading to signifi(cid:173)
`cantly increased small bowel mass'. The PGDP that exhibited the
`greatest intestinal growth-promoting activity in a murine assay was
`identified as glucagon-like peptide 2 (ref. 8), a 33-amino acid pep(cid:173)
`tide hberated in the enteroendocrine L cells of the small and large
`bowel.
`GLP-2 administered subcutaneously to mice for 10- 14 days
`induced a 1.5-2-fold increase of the mass of the small bowel.
`Comparable doses of GLP-2 (on a per weight basis) administered
`to rats did not result in an increase in small bowel weight. As all of
`these experiments were carried out with rat GLP-2, species-specific
`differences in the biological activities of these peptides did not
`account for the differential bioactivity in mice versus rats. We show
`·that GLP-2 is inactivated by the enzyme DPP-IV (EC 3.4.14.5) fol(cid:173)
`lowing N-terminal cleavage at the position 2 alanine, and that the
`DPP-IV-mediated peptide inactivation is a major determinant of
`the intestinotrophic activity ofGLP-2 in vivo.
`
`Results
`GLP-2 administration to mice stimulates crypt cell proliferation
`leading to increased mass of the intestinal mucosa'. Administration
`of rat GLP-2 via daily subcutaneous injection to mice in doses of
`10-200 µg/kg consistently results in the induction of increased
`small bowel mass after 10-14 days .... To ascertain wh ether GLP-2
`would exhibit similar growth-promoting properties in rats, we·
`administered rat GLP-2 (either 2.5 or 25 µg in a 10% gelatin for(cid:173)
`mulation subcutaneously twice daily for 10 days) to 200 to 250
`gram Fis,her 344 rats. GLP-2 treatment was associated with an
`
`673
`
`CFAD Exhibit 1021
`
`1
`
`

`

`© 1997 Nature Publishing Group http://www.nature.com/naturebiotechnology
`
`100
`
`Figure 1. Effect of GLP-2 In wlld·type Fischer rats. Rat GLP·2 (2.5 •
`or 2& µ,g • in 10% gelatin) or 10% gellltln alone (control • I was
`admlnldenid 8Ubcutaneoualy, every 12 houra for 10 day• to male
`Flacher rats (2.00-25o grame, n = 3 per group). The vill1.11 and crypt
`height wu a111111 d by mlc:roscopy from multiple sections (n = 10
`eectlom per rm) of proximal jejunum (PJ), dilltal Jejunum (DJ), and
`dlltal Ileum (DI). " "' pc:.05.
`
`increase in villus height iD the proximal and distal jejunum
`(p<0.05) but not in the ileum (Fig. 1). A small increase in crypt
`depth was also detected in the jejunum following GLP-2 treat(cid:173)
`ment. In contrast to the increase in mucosa! epithelial height no
`change in small bowel weight was observed following treatment
`with either dose of GLP-2 (Fig. 1).
`The difference in the effect of GLP-2 on the height of the
`intestinal mucosal epithelium compared to the small bowel mass
`in normal rats suggested· that GLP-2 may be considerably less
`intestinotrophic in rats compared to mice•. Structurally related
`members of the glucagon peptide superfamily such as glucose(cid:173)
`dependent insulinotropic polypeptide (GIP) and glucagon-like
`peptide 1 (GLP-1) contain a penultimate N-terminal alanine
`residue and are inactivated by the enzyme DPP-IV10•u.
`
`Nevertheless, GIP and GLP-1 are biologically active when adminis(cid:173)
`tered to rats. One explanation for the markedly reduced biological
`activity of GLP-2 in rats may be due to species-specific differences
`in the inactivation of GLP-2 by DPP-IV, as GLP-2, like GIP and
`GLP-1, also contains a His-Ala dipeptide sequence at the N-termi(cid:173)
`nus.
`Native rat GLP-2 was incubated with huinan placental DPP-'IV
`for 24 hours at 37"C, following which peptides were examined by
`HPLC. Incubation of peptide in the absence ofDPP-IV was not
`associated with the generation of significant amounts of detectable
`· cleayage products (Fig. 2A). In contrast, analysis of rat GLP-2 fol(cid:173)
`lowing incubation with DPP-IV demonstrated the presence of a
`second major peak (Fig. 2B) that eluted in the identical position as
`synthetic GLP-2(3-33). The identity of the peak as GLP-2(3-33)
`was confirmed by a combination of mass spectrometry and amino
`acid analysis. To determine ifGLP-2 undergoes a .similar DPP-IV(cid:173)
`mediated cleavage to GLP-2(3-33) iii rat plasma, we incubated rat
`GLP-2 for 24 hours with serum from wild-type rats or DPP-IV(-)
`rats. The HPLC profile of immunoreactive rat GLP-2 following
`incubation with wild-type rat serum showed a major peak that
`eluted in the identical position as synthetic GLP-2(1-33) (Fig. 3A).
`A second less abundant, but clearly detectable peak was observed
`that eluted in the identical position of synthetic GLP-2(3-33). In
`contrast, following incubation of rat GLP-2 with serum from a
`DPP-IV(-) rat, minimal amoWlts of immunoreactive GLP-2(3-33)
`were detected (Fig. 3B).
`.
`To test the hypofhesis that DPP· IV is a major determinant of
`GLP-2 activity in vivo, we examined the biological effects ofGLP-2
`in DPP-IV(-) rats. These rats, derived from a normal colony of
`Fischer 344 rats, exhibit decreased DPP-IV activity in vivo due to
`translation of a mutant form of the protein with decreased biologi(cid:173)
`cal activity',,". In contrast to the res~ts obtained following GLP-2
`administration to wild-type Fischer rats (Fig. 1), rat GLP-2 (25 µg
`twice a day) induced a statistically significant increase in both the
`height of the mucosal epithelium and the mass of the small bowel
`(Fig. 4, p<0.05-0.01 and p<0.0001, GLP-2 Vs. control, for mucosal
`· height and small bowel weight, respectively). These data demon-
`
`a
`
`B
`
`90.00 ....... - - - - - - - - - - - - - -
`b
`
`80.00
`
`'70.00
`
`~ 40.00
`
`so.oo
`
`a
`
`A
`
`l!I00.00
`
`250.00
`
`200.00
`
`150.00
`
`100.00
`
`150.00
`
`20.0
`
`30.0
`Retention time (min)
`
`20.0
`
`30.0
`Rettntion time (min)
`
`Figure 2. HPLC chromatogram of n.W. rat OLP-2 MfoN (A) and aft« (B) ~ent with DPP·IY. Incubation with human placental DPP-IV was
`.t 37°C for 24 houra. The peaks ma1<ed a Md b COINSpond to the ~on positions of synthetic rat GLP·2(1·3S) and GLP-2(3-33), reepectlvely.
`The Identities of peeks a and b were confirmed by mllll8 apec:trometry and amino acid aequel1Q9 analpie.
`
`874
`
`NATURE BIOTECHNOLOGY VOLUME 15 JULY 1997
`
`2
`
`

`

`© 1997 Nature Publishing Group http://www.nature.com/naturebiotechnology
`
`AESmAllCH
`
`A
`
`60
`
`ab
`l l
`
`B
`
`60
`
`. 40
`~
`D.
`..I 30
`<?
`!: 20
`t:
`
`..
`
`10
`
`·C\I
`
`I
`
`40
`~ 30
`..
`"
`!! 20
`r:
`
`10
`
`c
`
`50
`
`N 40
`I
`D.
`
`a 30
`..
`!!: 20
`
`t: 10
`
`SD
`
`60
`
`70
`
`eo
`
`90
`
`60
`
`60
`
`70
`
`BO
`
`90
`
`70
`60
`Fractions
`F"191n 3. Amt OLP•2(1-38} CA *Id B) or r(Gly2JGLP.2 (0.1 µ.g peptide) wH incubated In eeparata experiments In MrUm from contn:ll rllla CA and CJ
`or l)pp..IV(·) rm (B), for 24 hours. PeptidM were then eepareted by HPLC and analyzed by RIA. Peelm a and b lndicatle the elution positions of
`aynttmic rat GLP-2(1-33) end GLP.2{3'-33), ,...,,.ctlvely. Dela ahown 19 ...,...entlltlve of n = 5 experimenta.
`
`so
`
`80
`
`90
`
`strate significantly increased intestinotrophic activity of GLP-2 in
`DPP-IV(-) compared to wild-type rats and are consistent with the
`hypothesis that expression of the enzyme DPP-IV in wild-type rats
`is critically important for inactivation of GLP-2 and hence modu(cid:173)
`lation of its bioactivity in vivo.
`The demonstration that the enzymatic inactivation of GLP-2
`may be a limiting deter~ant of GLP-2 action prompted us to
`determine whether amino acid substitutions that confer resistance
`to DPP-IV cleavage might render the GLP-2 molecule more bioac(cid:173)
`tive in wild-type rats in vivo. A rat GLP-2 analog containing a
`glycine for alanine substitution at position 2 was synthes.i7.ed and
`incubated with human DPP-IV in vitro. No detectable cleavage of
`r[Gly2)GLP-2 to ~LP-2(3-33) was observed (Fig. 5), consistent
`with the notion that the penultimate alanine is a major determi(cid:173)
`nant ofDPP-IV recognition and cleavage ... Similarly, incubation
`of r[Gly2]GLP-2 with rat serum followed by HPLC and RIA
`demonstrated minimal cleavage to GLP-2(3-33) in vitro (Fig. 3C) ..
`Taken together, th~ observations demonstrate that r[Gly2]GLP-2
`is resistant to OPP-IV-mediated cleavage and as a result. raise the
`possiblity that this analog may be comparatively more bioactive
`than the native peptide in vivo.
`To determine the intestinotrophic properties of r[Gly2]GLP-2
`in vivo, wild-type and DPP-IV(-) Fischer rats were treated for 10
`days with twice daily injections of25 µg of r[Gly2]GLP-2, follow(cid:173)
`ing which the rats were sacrificed for analysis of small bowel
`weight and histology. In contrast to the negative results obtained
`with the native rat peptide, r[ Gly2] GLP-2 treatment of wild-type
`rats was associated with a significant increase in both villus height
`(p<0.05-0.0001) and small bowel mass (p<.0001) (Fig. 6).
`r[Gly2]GLP-2 was also slightly more intestinotrophic (as assessed
`by analysis of small bowel weight and histology) than native rat
`GLP-2 in DPP-IV(-) rats (Fig. 4). The combination of in vitro and
`in vivo analyses suggests that DPP-IV is a major determinant of
`GLP-2 bioactivity.
`
`Diacuaion
`The glucagon peptide superfamily comprises a number of peptides
`that contain an alanine residue at position 2, consistent with the
`known recognition cleavage sites (alanine or proline) for the pro(cid:173)
`tea8e OPP-IV. As the majority of these peptides require an intact
`N-terminus fur full biological activity, cleavage of the peptides at · ·
`the N-terminus generally inactivates the biological activity in vivo.
`For, example, native growth hormone-releasing hormone (l-44) or
`(1-40) is rapidly inactivated by DPP-IV to generate the GRH(3-44)
`or GRH(3-40} peptides. Hence, a major component of plasma
`im.munoreactive GRH is in fact not the mature intact peptide but
`the N-terminally degraded, biologically inactive forms that exhibit
`
`BOO
`
`70
`
`..
`
`.
`. . -<19
`-Af
`.a1
`
`-6~
`
`~]
`i-1"'
`
`- ~ -
`
`-
`
`CRYPT
`Figure 4. GLP-2 lnducu small bowel growth in DPP-IV(·) l'llta. DPP(cid:173)
`IV(·) male Racherms (-a!HI& aram-. n m3per~)w.rehellld with
`8llllne elone (control II), or rat ~2 •·or l(Oly2JGLP·2 • 25 µ.g, twice
`mlly for 10 mys.· Hi.tologlcel ~ of the lrNll bOW9I proxlnwI
`Jeft,nan (PJ), dlatal jejunum (DJ), md dletal Plum (Df) Wll8 cainled out
`.. deecrlbed In Exper'.mental protocol. *pc0.05, **p<0.01, -P<().0001
`
`a longer circulating t 112 in vivo15
`• GRH analogs that can be synthe(cid:173)
`sized are resistant to DPP-IV and exhibit greater stability and a
`longer half-life in vivo'$-''.
`Circulating forms of GLP-1 are also rapidly cleaved by DPP-rV
`to generate GLP-1(9-36) amide (and possibly GLP-1(9-37}}, pep(cid:173)
`tides that are biologically inactive in vivo1°'11
`• The observation that
`11
`•
`GLP-1 (9-36) amide may function as an antagonist at the GLP-1
`receptor'' suggests that DPP-IV may play an important role in
`determining the ratio of circulating agonist/antagonist in vivo, and
`hence may be a key determinant regulating signalling at the level of
`the GLP-1 receptor.
`The relative degree of peptide degradation by DPP-IV appears
`to be substantially greater for GLP-1 than for GLP-2. More than
`60% of GLP-1 was cleaved a1tcr only a 30 minute incubation with
`human plasma at 37°C". Consistent with in vitro data demonstrat(cid:173)
`.ing the rapid degradation of GLP-1 at the N-tenninus, a substan(cid:173)
`tial proportion of circulating immunoreactive GLP-1 in human
`pla,sma from fasting individuals is actually the N-terminally tnm(cid:173)
`cated GLP-1(9-36) amide, which represents the majority of
`immunoreactive GLP-1 in the postprandial ~tate11• In contrast, we
`observed less than 10% degradation ofGLP-2 after a much longer
`24-hour incubation with rat plasma in vitro. Similarly, the major
`molecular form detected after an 8-hour incubation of [us -I) GLP(cid:173)
`l (7-36) amide with OPP-IV was the N-terminally cleaved ["'(cid:173)
`I]GLP-1(9-36) amide10 whereas the major form ofGLP-2 detected
`
`NA1URE BIOTECHNOLOGY VOLUME 15 JULY 1997
`
`875
`
`3
`
`

`

`REllEARCH
`
`A
`
`0. 3000
`
`0 .2500
`
`0 .2000
`
`=a0.1500
`
`0 .1000
`
`o.osoo
`
`0.0000
`
`B
`
`0.400
`
`0.300
`
`i
`
`0.200
`
`0.100
`
`o.ooo
`
`20.0
`
`Retention time (min)
`
`30.0
`
`a
`
`20.0
`
`Rdmltion time (min)
`
`30.0
`
`0 and after
`F'19ure S. HPLC ~ of.r(Gly2]GLP-2 beb9 (A)
`(B) trMbnent wtlh DPP.IV at S7°C for 24 hour&. 1he puk marked"
`COft'MPOnda tD the elution polition of avnthetlc r(Gly2JGLP-2.. The
`ldentlly of peak ·a was confirmed by mass apectrometry and amino
`llCld Mquenea analysis.
`
`after a 24-hour incubation with DPP-N was the intact 33-amino
`acid peptide. These data imply that the relative stability of GLP-2
`in vivo may be somewhat greater than that observed for GLP-1.
`The importance of DPP-IV for the control of GLP-1 degrada(cid:173)
`tion is supported by studies demonstrating reduced N-terminal
`degradation of GLP-1 following infusion of GLP-1 into the DPP(cid:173)
`
`IV-deficient rat10• The OPP-IV-deficient rat, derived from a
`
`Japanese strain of Fischer 344 rats1i, contains mRNA transcripts
`for DPP-N but reduced levels ofDPP-N protein and enzymatic
`activity12-""0 due to translation of abnormal isoforms that fail to be
`processed to the biologically active mature glycosylated enzyme21
`•
`Although our in vitro data suggests that GLP-2 may be relatively
`less susceptible to degradation by DPP-IV compared to GLP-1,
`analysis of the molecular forms of GLP-2 in rat plasma demon(cid:173)
`strate a considerable amount of circulating GLP-2(3-33), consis(cid:173)
`tent with DPP-IV-mediated cleavage of this peptide in vivo (data
`
`878
`
`© 1997 Nature Publishing Group http://www.nature.com/naturebiotechnology
`
`a
`
`600
`
`... J:
`
`~ 30
`w
`J:
`
`20
`
`VILLI
`
`CRYPT
`
`Figure 8. Effect of r(Gly2JGLP-2 on rat small bowel growth. Male Fis(cid:173)
`cher rats (~60 grams at tho start of the experiment, n = 3 per group)
`in o.s ml aallne subcuta(cid:173)
`were Injected with 20 µ.g r{Gly2JGLP-2 •
`neoully twice dally (or aallne alone, control Im) for 10 daya, Hiatologi(cid:173)
`cal analyala of the small bowel proximal jejunum (PJ), dlatal jejunum
`(DJ), and dlatal Ileum (DI) was carried out as described in Experimen(cid:173)
`tal protocol. •P<().05, -p<0.0001
`
`not shown}. Consistent with the importance of DPP-N for the
`inactivation of GLP-2, we did not observe signiticant induction of
`~mall bowel growth following twice daily administration ofGLP-2
`to normal rats. Administration of pharmacological amounts of
`GLP-2 to rats may not necessarily be associated with increased
`concentrations of biologically active intact ·peptide, perhaps due to
`relatively increased OPP-IV activity in the rat. In contrast, treat(cid:173)
`ment of DPP-IV-deficient rats with unmodified rat GLP-2 was
`associated with a statistically significant increase in small bowel
`mass, providing clear evidence in support of a key role for DPP-rv
`in the regulation of GLP-2 activity in vivo.
`The serine protease DPP-IV is expressed in a broad range of tis(cid:173)
`
`sues and at very high levels in the kidnef1• DPP-IV activity has
`also been localized to the small intestine of both rats and humans,
`with the highest levels detected in the jejunum and ileum'..,.,
`Although the precise cellular target for GLP-2 ac:tMty has not been
`'identified, the demonstration that GLP-2 stimulates crypt cell pro(cid:173)
`liferation suggests that GLP-2 may exert its effects in part via a
`direct effect on crypt stem cells. Intriguingly, a crypt to villus gra(cid:173)
`dient of DPP-N has been described in the rat jejunum, with low
`levels ofDPP-IV in crypt cells increasing progressively in differen(cid:173)
`tiated enterocytes:zs. Local region-specific differences in intestinal
`DPP-IV expression may modulate the activity of GLP-2 by prefer(cid:173)
`entially inactivating the peptide at different sites along the crypt to
`villus axis.
`The GLP-2-mediated increase in villus height, but not small
`bowel weight, suggests that the amount of biologically active circu(cid:173)
`lating GLP-2 in these experiments was slightly below the threshold
`required to induce changes in small bowel mass. Consistent with
`this hypothesis, we have observed that mice treated with a low dose
`of GLP-2, 250 ng every 12 hours, exhibit a statistically significant
`increase in jejunal crypt plus villus height.but not in small bowel
`weight'. In contrast, larger doses of GLP-2 increased both villus
`height and small bowel weight in mice in vivo'.
`The detection of GLP-2(3-33) as the major cleavage product
`generated following incubation of GLP-2 with DPP-IV has impli(cid:173)
`cations for interpreting 4ata from radioimmunoassays that employ
`antisera specific for the mid or carboxyterminal region of GLP-2.
`At the present time.little information is available about the circu(cid:173)
`lating forms ofGLP-2 immunoreactivity, however a single study
`
`NATURE BIOTECHNOLOGY VOLUME 15JULY 1997
`
`4
`
`

`

`• © 1997 Nature Publishing Group http://www.nature.com/naturebiotechnology
`
`RESEARCH
`
`has reported that GLP-2 levels rise following food ingestion in
`normal human volunteers". The antisera used for this study was
`raised against 34-amino acid synthetic human GLP-2
`(proglucagon 126-159), however the GLP-2 epitqpes recognized
`by these antisera have not yet been reported. Taken together, the
`data reported here further extend the central role of DPP-IV in the
`regwation of the ~iological activity of the glucagon-like _peptides
`and suggests that GLP-2 analogs exhibiting DPP-IV resistance may
`potentially be therapeutically useful in vivo.
`
`Statistical Analysi.9. Statistical significance was calculated by ANOVA using
`a SAS program (Statistical Analysis Systems, Cary, NC) for IBM computers.
`
`Acknowledgements
`This work was supported in part by grants from the Medical Research Council
`of Canada (PLB ), the National Science and Engineering Research Council of
`Canada (OJD) and Alie/ix Biophannaceuricals Inc (D]D). DJD.is a consultant
`to Alie/ix Biophannaceuticals Inc. T/1e authors tl1ank Karim Meghji and
`Susanna Ng for expert rechnical assistance.
`
`Experimental protocol
`All chemicals were from Sigma Chemicals (StLouis, MO) or Baxter Travenol
`Canada (Toronto, Ontario). Rat GLP-2 was obtained from the American
`Peptide Company (Sunnyvale, CA) or from California Peptide Research
`(Napa, CA) and·was greater than 99% pure 1,y HPLC. Rat (Gly2)GLP.-2(1-
`33) and rat GLP-2(3-33) were obtained from California Peptide Research
`and were 97% pure by HPLC. Diprotin A was purchased from Novabiochem
`(San Diego, CA) or Sigma. Hwnan placental dipeptidyl peptidase-IV (spe(cid:173)
`cific activity 5,000 mU/mg protein) was purchased from Calbiochem (San
`Diego, CA). Th!sylol was obtained from Bayer (Toronto, Ontario) and ED.TA
`was from ACP Chemfotl (Montreal. Quebec).
`.AnimUs. Fischer rats were obtained from Charles River Laboratory (St.
`Constant, Quebec). A small colony of DPP-IV negative rats,." was estab(cid:173)
`lished following proctUe:ment of the animals from Ray Pederson, University
`of British Columbia (Vancouver, ~C). The rats were not restricted by diet or
`activity during the experiment and were housed 2-3/cage under a lighf/dark
`cycle of 12 h. Animals were fasted overnight ( 12-14 h) prior to sacrifice.
`AninW and Twue ~g. Rats were anesthetized with CO, and the
`small intestine was rcmoveli from the peritoneal cavity (from pylorus to
`cecum), cleal)ed, weighed and meas.ured. For comparative purposes, sec(cid:173)
`tions from each animal were obtained from similar anatomical positions,
`-16, 38, and 65 an.from the gastroduodenal junction for proximal and dis(cid:173)
`tal jejunum and ileum, respectively. Each small bowel fragu,lcnt was opened
`longitudinally on ·its ·antimesenteric border, sectioned, and then placed in
`10% formalin (v/v) overnight.
`Histological sections 5 µm thick were cut. stained with hematoxylin and
`eosin and u~d for.micrometry and morphometric analym as previously
`described'. Intestinal miaometry Wllll performed ~g a Leitz (Wetzar,
`Germany) microscope with a video camera connected to a computer moni(cid:173)
`tor. The microscope was cali'brate~ at 4X, lOX, ~d 25X magnification and ·
`the: same microscope was-used for all evaluations. Crypt and villus height
`was measure<l by enmining at least IO longitudinally orienred villi from
`each slide for promnal and distal jejunum and distal ileum and is expressed
`inµm.±SEM.
`.
`· DPP-IV deavap of rat GLP-21111d r[Glf2]Gi.P-2. Peptide solutions were
`prep°arcd at a concentration of 0.2 mg/ml in 40 mM PBS, pH 6.5. A 2.5 µI
`aliquot (0.125 mU) of human DPP-IVwas added tO each 50 µI aliquot (IO
`µg) of peptide s'olution, and the mixtures were incubated at 37"C for 24 h.
`· The enzyme incubations \m'e qiienched by addition of 50 µI (200 µg) of a 4
`mgfml soiution ofDip,rotin A in PBS; Following in~batfon of peptides with
`DPP-IV in vitro, a 100 µI aliquofof the quen(:bed incubation mixture was
`injected onto a Rainin Dynamax Cl.8 (250X4.6 mm) column. The samples
`were duted using a linear solvcnt·gradient [3~0% B; solvent A = 0.1 %
`(vlv) TFA in water and solveil.t B = 0.1% TFA in acetonitrile) at a flow rate of
`l ml/min. Blution was monitottd by absorbance at 214 nm.
`Blood for in vitro incubations was collected from. wild-type :WJStar and
`Fischer rats, following which peptides {O.l µg) were incubated with l ml of
`serum fur 24 hat 31"C. The reaction was tl:nninmd by addition of 110 Jil TED
`('Ikasylol:EDTA:Diprotin Ao [5000 KIU/ml:l.2 mgfml:O.I mM). The sample
`was thenmixied with 2mlofof1% (vf.v) TFA. pH adju:stzd to 2.5 with diethy(cid:173)
`lamine and applied to a C,. Sep Pak cartridge (Waters AsSociates, Bedford,
`MA). The peptides were eluted with 3.0 ml 8096 (v/v) isopropniol in 0.196
`(v/v) TFA and stnttld at -70 C prior to radioimmunoassay. Peptides were sepa(cid:173)
`rated by HPLC as desCribed.above, using a C,. uBondapak column (Waters)
`with a fiowrate of 1.5 ml/min. Fractions wm:mlkdMcw:ry0.3 min.
`GLP-2 RadioimnnmOa-r-RIA.for GLP-2 was carried out Using an anti(cid:173)
`serum (UlTH-7) tbiit recognius the midsequence of GLP-2 (amino acids
`25-30) and cross-reacts equally with GLP-2(1-33} and GLP-2(3-33) (data
`not shown). The detection -limits of the assay were 10-2000 pg/tube; the ·
`· intra and interassayvariations were 3.6% and 8.3%, respectively.
`
`1. Taylor, R.G. and Fuller, P.J. 1994. Humeral regulation of Intestinal adaptation.
`Bllilli«esC!ir> .. Encioaino/. andMtttBb. 8:165-183.
`.
`2. Gleeson, M.H .• Bloom, S.R.. Polak, J.M., Henry, K., and Dowling, R.H. 1971.
`Endocrine tumour In kidney effecting smell bowel structure, motifity, and absorptive
`functlon. Gut 12:nS.-782.
`.
`3. Stevens, F.M., Flanagan, R.W., O'Gonnan, o.. and Buchanan. K.O. 1984.
`Glucagonomasyndroma demoristratJng giant duodlinaJ vi111. Gut 25:784-791.
`4. Bloom, S.R. and Polak, J.M. 1982. The hormonal 1>;11ttem of intestinal adaptation (a
`major role for enteroglucagon). Scand. J. GaslnlenteroJ. 1~1 03.
`5. Fuller, P.J., Beveridge, 0.J., ahd Taylor, R.G. 1993. 119111 proglucagort gene
`expression in the rat: characterization in lntestlnal adaptation using in. situ
`hybridlzation:Gastroenterolog104:459-466.
`6. Bestennan, H.$., Adrian, T.E., Malllnson, C.N .. Christafides, N.O., Sar.ion, D.L,
`Pera, A. et al, · 1982. Gut hormone l81euse after intestinal rosection. Gut
`23:854-e61:
`7. Ehrlich, P., Tucker, D., Asa, S.L, Bnblker, P.L, and Drucker, D.J. 1994. lnhlbitionof
`pancreatic proglucagon gene expression In mice bearing subcutaneous endocrine
`tumors. Am. J. Physlol. Endocrine/. Melab. 287:E662-E671.
`8. Drucker, D.J .. Ehrlich, P., Asa, S.L;and Brubaker, P.L 1996. Induction of Intestinal
`epithelial proliferation by glucagon-Jlke pept1d!I 2. Proc. Natl. AC/Id. Sci. USA
`113:7911-7916.
`.
`9. Tsai, C.-H., Hill, M., and . Drucker, O.J. 1997. Blologlcal determinants of
`intestinotrophlc properties of GLP-2 in vivo. Am. J. Physlol. 272:G662-G668:
`10. Kle1fer, T.J., Mclntoslt, C.H.S., and Pederson, RA 1995. Degradation of glucose(cid:173)
`dependent lil&ulinotroplc polypeptide andtrunceted gl~-llke pep1ide 1 In vitro
`and In vivo bydipeptfdyt peptldas& IV. Endoctil1cJogy 136:~ .
`11. Deacon, C.F.. Johnson, A.H., and Holst, J.J. 1995. Oegradatlo.n of glucagon-like
`peptid&-1 l;>y human ~ In vitro yields an N-tei:nilnally lr!Jncatad pGptldd that is
`e m!IJorendogenous metabolite in vivo. J. Cl(r>. En<JocriflOI. Mefab:80:952-957 . .
`12. watanabe, Y., Kojima. T., end Fujimoto, Y. 1987. Oellciency of membrane-bound
`dlpeptidyl aminopeptfdese JV in a certain rat strain. Experlent/a43:40G-401.
`13. Thompson, N.L, Hbcson, o.c .. Callanan, H., Panzica. M .. Fl8nagan, D., Faris, R.A,,
`et al •. 1991. A~ rat sub&1nlil deficient In dlpePtidyl peptldas& JV activity
`makes normal steady-state RNA levels and an altered protein. 8/ochem. J.
`273-.497-602.
`14. Bong1119, J., Lambros, T., Ahmad, M., llf\d Helmer, E.P. 1992. Kinetics of dlpeptidyl
`pep1idase JV proteolysis of growth hormone-releasing factor lind analogs. Biochlm.
`8lophys. Acta1122:14].-153.
`15. Frohman, LA, Downs, T.R., Heimer, E.P., and Felix, A.M. 1989. Dipeptidyl
`peptidase N and· trypslrHlke enzymatic degradation of human growth honnone(cid:173)
`releaslng hcrmone In pllllllllB. J. Clln. /nV611t. 83;1533-1640.
`16. Lance, VA, MU!phy. WA, Suelras-Qlaz, J., and Coy, D.H. 1984. SL1psr-active
`analogues of growth hormone-ralaaslng factor (1 ·29)-amlde. B/dchem. Blophys.
`Fm. commun.1111:2~272.
`17. Frohman, LA., Downs, T.R., Wlfflams, T.C., Helmer, E.P., Pan, Y.-C.E., and Fellx,
`AM. 1988. Rapid enzymatic degradatlon of growth bonnone raleaslng factor by
`plasma In vivo and in vivo to a blologlcally Inactive product ·cleaved at the NH,-
`tenninuss. J. Clln. Invest. 78:9CMHl13.
`.
`18. Mentleln, R., GelwttZ. B., and Sctimlclt, W.E. 1993. Dlpeptldyl-peptldase N
`hydrolys!JS gastric Inhibitory polypept!de, glucilgon-llke peptlde·1(7-36) amide,
`peptide hlstld!ne methionine. and Is responsible for their degradation In human
`serum. Eur. J. Blochem. 214:829-835.
`19. Knudsen, LB. and Pr1dal, L 1996. Glucagon·Wke peptlde-1-{9-36)amfde ls e major
`melllbollte of glucagon-Rke peptlde-1-(7-36) amide after Jn vivo admlnlstratlon to
`dogs and it acts as en antagonist on the pancreatic receptor. Eur. J. Phsrmacol.
`318:429-435 . .
`20. Thompson, N:L, Hixson, D.C., Cellanan,.H., PBl')zlca, M., Flanagen, D., Faris, R.A:.,
`et aJ. 1991. A Fischer rat substraln deflc:Jent 1(1 dlpoptldyl peptldaae IV activity
`makes noonaJ ataady-stala RNA levels and an altered protein. Use as a liver-cell
`trlWlSplantlltl model 81ocherri. J. 273:491-502 • .
`. 21. Erickson, R.H., Suzuki, Y.; sedlrnayer, A, and Kim, Y.S. 1992. Blosynthesls and .
`~of altlll8d immalut8 fonns of ln!Mtinal dlpeptldyl peptidase IV in a rat
`slJ8ln lackingtheenzyme..L Biol. Chem. ~:21~1629.
`22. HJdebrandt, M., Reutter, W., and Gitlin, J.D. 1991: Tlosue-epeclflo regulation of
`IV expn!&Sion during development . Biochern. J.
`dlpeptldyl peplidas&
`277:331-334.
`2S. Darmoul, D., Voisin, T. Couvlneau, A...Rouyet-fessmd, C., Salomon, R., Wang, Y.,
`et el. 1994. Regional expiesslon of epithelial cfipeptidyl peptidase IV Iii the human
`kllesmes. Blochem. Biophys. Ras. Commun. 20S:1224-1229.
`24. Sal, JP.F. 1994. OIStrlllulion of brush-l>order membrane peplld.ases along the rat
`fntestine.Ph&m;I. Res. 11:89.7...QOO.
`25. Oarrnoul, D., Rouyer-Fessald, C., Blais, A., Voisin, T., Sllpln, C., Baricault. L, et al.
`1991. Dipeptidy1 peptidase JV exp1868lon In rat Jejune! crypt-villus axis Is controlled
`at mRNA level.Am, J. Physiol. 261 :G763--0769.
`·
`26. Orskov, C. and Holst, J.J. 1967. Radio-immunoassays for glucagon·llke peptides 1
`and 2 (GLP-1 and GLP-2). Scand. J. Olin. Lab. Invest. 47:1~174.
`
`NATURE BIOTECHNOLOGY VOLUME 15 JULY 1997
`
`677
`
`5
`
`