printer-friendly

Sample Business Contracts

License Agreement - Biogen Inc. and Tanox Biosystems Inc.

Sponsored Links


                                LICENSE AGREEMENT

   This License Agreement (hereinafter "Agreement") is made and entered into as
of the 1st day of June (the "Effective Date") between Biogen, Inc., a
Massachusetts corporation having a principal place of business at 14 Cambridge
Center, Cambridge MA, 02142 ("Biogen") and Tanox Biosystems, Inc., a Texas
corporation, having a principal place of business at 10301 Stella Link, Houston,
TX 77025 ("Tanox").

   NOW THEREFORE, the parties hereby agree as follows:

BACKGROUND

   Biogen owns certain patent rights identified in Appendix A hereto, relating
to ANTI-CD4 MONOCLONAL ANTIBODIES.

   Biogen has additional rights to technical data and information pertaining to
anti-CD4 monoclonal antibodies.

   Tanox wishes to be licensed under certain of the TECHNOLOGY (as defined
below) and patent rights and Biogen is willing to grant Tanox such a license,
for the consideration and under the terms set forth in this Agreement.

1. DEFINITIONS

1.1    "AFFILIATE", as applied to either party, shall mean any corporation,
       firm, partnership or other entity which directly or indirectly owns, is
       owned by, or is under common control with, a party to the Agreement to
       the extent of at least fifty percent (50%) of the equity or other
       ownership interests (or such lesser percentage which is the maximum
       allowed to be owned by a foreign corporation in a particular
       jurisdiction) having the power to vote on, or direct the affairs of, the
       entity.

                                       1
<PAGE>
1.2    "DISTRIBUTOR(S)" shall mean a person or entity in a country who buys
       LICENSED PRODUCT from Tanox or its AFFILIATES or SUBLICENSEES and who,
       under an implied license, sells such LICENSED PRODUCT in that country.

1.3    "FIELD" shall mean the manufacture, use, importation, offer for sale or
       sale of LICENSED PRODUCT for human use only, including therapeutic,
       prophylactic and diagnostic use.

1.4    "FIRST COMMERCIAL SALE" shall mean in each country of the TERRITORY, the
       first sale of a LICENSED PRODUCT by Tanox or any of its AFFILIATES,
       DISTRIBUTOR(S) or its SUBLICENSEES to a third party in connection with
       the nationwide introduction of LICENSED PRODUCT by Tanox or any of its
       AFFILIATES, DISTRIBUTOR(S) or its SUBLICENSEES following marketing and/or
       pricing approval by the appropriate governmental agency for the country
       in which the sale is made. When governmental approval is not required or
       when sales can be made through named patient sales (also including
       "compassionate use sales", "treatment INDs" or their equivalents) prior
       to governmental approval, such FIRST COMMERCIAL SALE in that country
       shall be the first sale in connection with the nationwide introduction of
       LICENSED PRODUCT.

1.5    "IND" shall mean an Investigational New Drug Application filed with the
       U.S. Food and Drug Administration ("FDA") or an equivalent filing in the
       TERRITORY.

1.6    "LICENSED PRODUCT(S)" shall mean any composition(s) (i) the manufacture,
       use, importation, offer for sale, or sale of which, but for this license,
       infringes a VALID CLAIM of one or more PATENTS or (ii) which embodies any
       of the TECHNOLOGY.

                                       2
<PAGE>
1.7    "MAA" shall mean an application for regulatory approval to sell LICENSED
       PRODUCT in the European Union and is similar in purpose to an NDA in the
       United States.

1.8    "NDA" shall mean a New Drug Application or Biologic License Application
       ("BLA") or equivalent filed for LICENSED PRODUCT with the U.S. Food and
       Drug Administration ("FDA").

1.9    "NET SALES" shall mean the total gross invoice amount actually received
       by Tanox on the sale of LICENSED PRODUCT in the TERRITORY by Tanox and
       its AFFILIATES, SUBLICENSEES and DISTRIBUTOR(S) to third parties less the
       following items, as determined from the books and records of Tanox or its
       AFFILIATES, SUBLICENSEES or DISTRIBUTOR(S): (i) insurance and transport
       charges actually invoiced; (ii) amounts repaid or credited for rejection
       or return of LICENSED PRODUCT; (iii) sales or other excise taxes or other
       governmental charges levied on the invoiced amount and actually paid by
       the seller; (iv) custom duties and charges actually paid by the seller;
       (v) normal and customary trade and quantity discounts actually allowed
       and actually taken which relate to LICENSED PRODUCT.

      Sales of LICENSED PRODUCT between or among Tanox, its AFFILIATES,
DISTRIBUTOR(S) or the SUBLICENSEES shall be excluded from the computation of NET
SALES. Notwithstanding the previous sentence, the resale of the LICENSED PRODUCT
by the AFFILIATE, DISTRIBUTOR(S) or SUBLICENSEES to a third party who is not an
AFFILIATE, DISTRIBUTOR(S) or SUBLICENSEES of Tanox, shall be included in the
definition of NET SALES for the purposes of this Agreement.

      In the event that Tanox, its AFFILIATES, DISTRIBUTOR(S) or its
SUBLICENSEES receives in any transaction included within the definition of NET
SALES, any non-cash compensation or lower prices on other products in exchange
for any LICENSED PRODUCT, or sells LICENSED PRODUCT in other than an arms length


                                       3
<PAGE>
transaction, then the gross amount invoiced in such transaction shall be deemed
to be the gross amount that would have been paid had there been such a sale at
the average sale price of such LICENSED PRODUCT during the applicable royalty
reporting period in the country in which such disposition took place. The
preceding sentence shall not apply to the distribution at no cost of LICENSED
PRODUCT to physicians, hospitals or clinics for promotional purposes or academic
investigators for research and clinical trial purposes, which are not included
in the definition of NET SALES.

      In the event that LICENSED PRODUCT is sold in combination with one or more
other active ingredients, including one or more other antibodies having a
therapeutic effect (such antibody(ies) defined as an "Antibody Active
Ingredient"), the NET SALES on the combination product shall be calculated by
multiplying actual NET SALES of the combination product by the fraction obtained
after dividing A by the sum of A+B, in which "A" is the average selling price of
LICENSED PRODUCT sold separately by the same party during the same accounting
period in the country in which the sale of the combination product was made and
"B" is the average selling price of the other active ingredient(s) sold
separately by the same party during the same accounting period in the country in
which the sale of the combination product was made.

      If, however, no separate sales of LICENSED PRODUCT or Antibody Active
Ingredient(s) or other active ingredients are made then:

      A. If the combination product includes one or more Antibody Active
Ingredients (the combination product defined for the purpose of this sentence as
a "TRUE ANTIBODY COMBINATION PRODUCT"), the NET SALES of such TRUE ANTIBODY
COMBINATION PRODUCT shall be reduced by 50%; or

      B. If the combination product includes one or more active ingredients that
are not Antibody Active Ingredients, the parties shall negotiate in good faith a
mutually agreeable formula for determining appropriate reduction in NET SALES
with the goal of equitably determining the appropriate percentage value
represented by the components.

                                       4
<PAGE>
      The parties agree that in no event shall NET SALES of LICENSED PRODUCT or
any combination product of this Article 1.9 be reduced to less than fifty
percent (50%) of actual NET SALES of such LICENSED PRODUCT or combination
product by reason of any adjustment set forth in this Article 1.9.

      The parties further agree that if Tanox becomes (a) an AFFILIATE of; or
(b) merged with; or (c) acquired by: or (d) an acquirer of, a third party having
a market capitalization equal to, or greater than ten billion dollars ($10B) at
the time of such AFFILIATE formation or merger or acquisition, then NET SALES as
defined in the first sentence of this Article shall be redefined to mean the
total gross invoice amount (not dependent on whether the invoices were actually
paid) on the sale of LICENSED PRODUCT in the TERRITORY by Tanox and its
AFFILIATES, SUBLICENSEES and DISTRIBUTOR(S) to third parties, less the items
recited in the first sentence of this Article. All other provisions of the
definition of NET SALES shall remain unaffected by such AFFILIATE formation,
merger or acquisition.

1.10   "PATENTS" shall mean: (a) the patent applications and patents listed in
       Appendix A to this Agreement, any extensions, supplemental protection
       certificates, reissues, renewals, re-examinations, divisionals,
       continuations or continuations-in-part thereof, any foreign counterparts
       thereof, any patent issuing from any of the foregoing, which Biogen
       presently or hereafter owns and (b) such other patent applications and
       patents which Biogen presently or hereafter owns or controls that is
       required for the manufacture, use or sale of LICENSED PRODUCT.

1.11   "PROTEIN DESIGN LABS PATENTS" shall mean the patents and patent
       applications listed in Appendix B hereto.


                                       5
<PAGE>
1.12   "ROYALTY QUARTER" shall mean the three (3) months ending on the last day
       of March, June, September and December of each year.

1.13   "SUBLICENSEES" shall mean a third party (but not an AFFILIATE of Tanox)
       licensed by Tanox to make, use, import, offer for sale and sell LICENSED
       PRODUCT.

1.14   "TECHNOLOGY" shall mean any data, know-how, or other information, or any
       material, reagent or other substance relating to LICENSED PRODUCT or to
       anti-CD4 monoclonal antibodies which may be useful in the discovery,
       research, development, manufacture, use or sale of LICENSED PRODUCT or to
       anti-CD4 monoclonal antibodies and which is known to, and/or in the
       possession of, Biogen on the Effective Date and to which Biogen has a
       transferable right.

1.15   "TERRITORY" shall mean the entire world.

1.16   "VALID CLAIM" shall mean (i) any claim(s) in an issued, unexpired patent
       which has not been held unenforceable, unpatentable, or invalid by a
       court or other governmental agency of competent jurisdiction, in a
       decision that is unappealable or unappealed, within the time allowed for
       appeal, and which has not been abandoned or admitted to be invalid or
       unenforceable through reissue or disclaimer or (ii) a claim of a pending
       patent application which is pending as of the Effective Date and for
       which examination has been requested or, in the case of Japan and Canada,
       will be timely requested, and which claim shall not have been canceled,
       withdrawn, abandoned or rejected by an administrative agency from which
       no appeal can be taken. If in the TERRITORY there should be two or more
       such decisions conflicting with respect to the validity of the same
       claims, the decision of the higher or highest tribunal shall thereafter
       control. However, should the tribunals be of equal rank, then the
       decision or decisions upholding the claim shall prevail


                                       6
<PAGE>
       when the conflicting decisions are equal in number, and the majority of
       decisions shall prevail when conflicting decisions are unequal in number.

2.     LICENSES AND TECHNOLOGY TRANSFER

2.1    Biogen hereby grants Tanox a TERRITORY-wide, exclusive, royalty-bearing
       license under the TECHNOLOGY and PATENTS to make, have made, import, use,
       offer for sale, and sell LICENSED PRODUCT for use in the FIELD. As soon
       as reasonably possible following execution of this Agreement, Biogen will
       transfer all TECHNOLOGY listed on Appendix C that is in its control or
       possession to Tanox, and shall cooperate with Tanox in its initial use of
       the TECHNOLOGY to develop a LICENSED PRODUCT by making Biogen personnel
       with knowledge of the TECHNOLOGY available for initial meetings at Biogen
       as agreed, not to exceed two working days, and subsequent telephone
       conference(s) with Tanox during Biogen's normal business hours under the
       following conditions:

       (a) Biogen personnel are given at least reasonable advance notice of such
       conference(s);

       (b) Tanox shall direct all telephone calls to a Technology Liaison who
       will direct Tanox's attention to the appropriate person responsible for a
       particular TECHNOLOGY. Biogen shall provide Tanox with all available
       TECHNOLOGY within 3 months from the Effective Date of this Agreement;

       (c) Telephone conferences shall not exceed 5 hours per month for the
       first three months,

       (d) Any time spent by Biogen personnel on retrieval of information
       requested shall, except for information previously requested and not
       provided, be limited to the three month period after the Effective Date
       and shall be reasonably based upon pre-existing commitments and the time
       required to complete the tasks requested by Tanox; and

       (e) Telephone conferences subsequent to the time period of subpart (b)
       (other than for pursuing information or TECHNOLOGY previously requested
       but not provided) shall not exceed 1 hour per month.


                                       7
<PAGE>
2.2    The term "exclusive" in Article 2.1 shall mean that Biogen cannot grant
       further licenses hereunder for LICENSED PRODUCT in the TERRITORY in the
       FIELD, subject only to:

      (a) United States Government rights, if any, that may be created by the
use of LICENSED PRODUCT by Dr. Keith Reimann, Beth Israel Hospital-RE-113,
Boston, MA 02215, pursuant to existing National Institute of Health ("NIH")
contract 1R01 HL 59747-01. Biogen warrants and represents that none of the
PATENTS were conceived and/or reduced to practice with any United States
Government funding and that neither Dr. Reimann nor any third party has any
rights to commercialize LICENSED PRODUCT. Biogen, upon Tanox's request, agrees
to provide Tanox with all data generated during the course of Dr. Reimann's NIH
contract (data generated as of December 1996 are attached as Exhibit D); and

      (b) Biogen's reserved right to use the TECHNOLOGY and PATENTS for internal
research and educational purposes.

2.3    The license granted to Tanox hereunder shall include the right to grant a
       sublicense to one SUBLICENSEE in each country. A copy of the sublicense
       shall be provided to Biogen. Tanox shall ensure that its AFFILIATE(S) and
       SUBLICENSEES to whom Tanox has extended or sublicensed its rights under
       this Article 2, shall comply with all of the terms of this Agreement to
       which Tanox is bound.

      Any sublicense granted by Tanox under this Agreement shall be subject and
subordinate to the terms and conditions of this Agreement except that:

      (a) the sublicense terms and conditions shall reflect that the
SUBLICENSEES shall not have the right to further sublicense;

      (b)the sublicense shall include a requirement that the SUBLICENSEES use
the same efforts to bring the subject matter of the sublicense into commercial
use as those efforts of Tanox under Article 9.1 of this Agreement; and

      (c) the sublicense shall expressly provide for the transfer of all
obligations, including the payment of royalties specified in the sublicense, to
Biogen or its designee, in the event that the present Agreement is terminated.


                                       8
<PAGE>
2.4    Except as expressly provided herein, nothing in this Agreement shall be
       deemed to grant either party any rights or license to any patent, patent
       application, technology, know-how or invention of the other party.

3. PAYMENTS AND REPORTING

   In consideration of the license rights granted to Tanox under this Agreement,
Tanox shall make the following payments to Biogen:

3.1                                    *

3.2    Tanox shall pay Biogen a royalty on NET SALES of LICENSED PRODUCT sold by
       Tanox, its AFFILIATES, SUBLICENSEES and DISTRIBUTOR(S) in the TERRITORY
       at the following rates:


    PORTION OF
    ANNUAL NET SALES                      ROYALTY RATE    OFFSET (ARTICLE 3.3)
    ----------------                      ------------    --------------------
           *                                    *                   *

                                       9
<PAGE>
3.3    If Tanox licenses PROTEIN DESIGN LABS PATENT(S) to make, have made, use,
       sell, offer for sale, or import LICENSED PRODUCT, then so long as a VALID
       CLAIM of a PROTEIN DESIGN LABS PATENT encompasses the LICENSED PRODUCT in
       the United States, the royalty otherwise payable to Biogen under Article
       3.2 on sales in the respective countries in which there are such VALID
       CLAIMS shall be reduced by the amounts set forth in the above Table of
       Article 3.2 under the heading "Offset" determined by reference to the
       amount of annual NET SALES in such countries.

3.4    As further consideration of the rights granted to Tanox under this
       Agreement, Tanox shall make the following nonrefundable, noncreditable
       payments to Biogen upon the first achievement of each of the following
       milestones.

   MILESTONE                                                    PAYMENT
   ---------                                                    -------
   Upon filing an IND or equivalent
   filing for first LICENSED PRODUCT                              *

   Commencement of Phase III Clinical
   Trial (or equivalent pivotal trial)
   of first LICENSED PRODUCT                                      *

   Filing BLA/MAA or equivalent filing
   for first LICENSED PRODUCT                                     *

   Regulatory approval anywhere
   in the TERRITORY of first LICENSED
   PRODUCT                                                        *

                                       10
<PAGE>
3.5    If, after executing this Agreement, Tanox becomes (a) an AFFILIATE of; or
       (b) merged with; or (c) acquired by: or (d) an acquirer of, a third party
       having an equal or greater number of employees than Biogen at the time of
       such AFFILIATE formation or merger or acquisition, then Tanox shall be
       obligated to make additional milestone payments in addition to those
       recited above in Article 3.4. Following such a merger or acquisition, the
       additional nonrefundable, noncreditable retroactive milestones are due
       thirty (30) days after the later of: (i) being triggered by the events
       described in the Table below or (ii) the date of the acquisition, or
       merger transaction and are as follows:

   MILESTONE                                                        PAYMENT
   ---------                                                        -------
   Commencement of Phase II

   Clinical Trial of first LICENSED PRODUCT                            *

   Commencement of Phase III
   Clinical Trial (or equivalent pivotal trial)
   of first LICENSED PRODUCT                                           *

   Regulatory approval by FDA of first LICENSED PRODUCT                *

   Regulatory approval by EMEA of first LICENSED PRODUCT               *


                                       11
<PAGE>
3.6    All royalties accrued pursuant to this Article 3 shall be paid to Biogen
       on a quarterly basis within forty five (45) days after the end of each
       calendar quarter with respect to NET SALES on sales of LICENSED PRODUCT
       made in such quarter.

3.7    Together with each royalty payment due under this Article 3, Tanox shall
       provide Biogen with a signed written statement certifying, separately for
       each type of LICENSED PRODUCT, the following information: (i) gross sales
       by Tanox, its AFFILIATES, SUBLICENSEES, and DISTRIBUTOR(S), by country;
       (ii) NET SALES separately by country and in total; (iii) quantity sold by
       Tanox, its AFFILIATES, SUBLICENSEES, and DISTRIBUTOR(S); and (iv) average
       sales price in each country. Tanox shall maintain, and shall use
       reasonable efforts to ensure that its AFFILIATES, SUBLICENSEE and
       DISTRIBUTOR(S) maintain, appropriate books of account and records of all
       sales of LICENSED PRODUCT in any calendar year for a period of three (3)
       full years after such calendar year. At Biogen's request, Tanox shall
       make such books of account and records available for inspection during
       normal business hours by independent public accountants appointed by
       Biogen for the purpose of verification of the amounts paid to Biogen
       under this Agreement. Biogen shall not conduct more than one audit in any
       calendar year.

   The cost of such audit shall be borne by BIOGEN unless it is established by
the audit that there has been an error which has caused Tanox to underpay by ten
percent (10%) or more for the period under audit, in which case the cost of such
audit shall be borne by Tanox. Tanox shall pay to Biogen any underpaid
compensation that is confirmed by the audit promptly and with interest at a rate
not to exceed the interest rate defined in Article 3.9 hereunder.

3.8    The amounts computed or specified under this Article 3 as due to Biogen
       are the actual amounts to be received by BIOGEN and shall not be reduced
       in any way, including but


                                       12
<PAGE>
       not limited to, withholding taxes and reduction by any liabilities
       incurred by Tanox or its SUBLICENSEES or DISTRIBUTOR(S) on Tanox's behalf
       (but not on Biogen's behalf) upon remittance to BIOGEN of the payments
       due hereunder, provided, however, that if required, Tanox shall be
       allowed to withhold taxes incurred by Biogen.

3.9    All payments made to either party hereunder shall be paid in U.S.
       Dollars. Monetary conversion from the currency of a foreign country into
       U.S. currency shall be made at the exchange rate in force on the last
       business day of the quarter in which the payment obligations were
       incurred as reported in The Wall Street Journal, or on such other basis
       as mutually agreed upon by both parties. Any amounts due under this
       Agreement that are not paid when due shall bear interest at the lesser of
       (i) *.

3.10   The obligation to pay royalties shall continue on a country-by-country
       basis from the date of first sale of LICENSED PRODUCT in that country
       until the later of (i) twelve (12) years from the date of FIRST
       COMMERCIAL SALE of such LICENSED PRODUCT in that country or (ii) the date
       on which the manufacture, use, sale, offer for sale or import of LICENSED
       PRODUCT is no longer covered by a VALID CLAIM of any PATENT in such
       country or in the country of manufacture.

4.   THE TANOX INTERFERON PROGRAM

As further consideration for the exclusive rights granted to Tanox hereunder,
Tanox agrees that it shall provide a written submission to Biogen of all IN VIVO
data using one or more Tanox interferon-Fc fusion proteins that demonstrate
"proof of concept" (such data to include, but not limited to, an animal model of
cancer or EAE). In furtherance thereof,


                                       13
<PAGE>
Tanox hereby grants to Biogen an exclusive option to obtain an exclusive,
worldwide license of Tanox's interferon-fusion proteins at commercially
reasonable terms and conditions, including royalties. Upon receipt of the Tanox
"proof of concept" submission, a 180 day period (the "Option Period") shall
commence within which Tanox and Biogen shall enter into good faith negotiations
for the exclusive license described in the previous sentence. If Tanox and
Biogen cannot mutually agree on terms of such a license during the Option
Period, then Tanox shall be free to license or otherwise transfer its rights in
any of its interferon-fusion proteins to a third party, provided that for a
period of 180 days after the end of the Option Period, Tanox shall not offer any
such third party terms for a license for an interferon-fusion protein that are
materially different from those last offered to Biogen, without first offering
such different terms to Biogen.

5. PATENT PROSECUTION AND LITIGATION

5.1    Biogen shall be solely responsible for prosecution and maintenance of the
       PATENTS. Tanox shall bear the cost of all matters relating to the
       maintenance and prosecution of the PATENTS, such costs to be creditable
       against any payments due Biogen. Biogen shall promptly notify Tanox of
       all information received by Biogen relating to the prosecution and
       maintenance of PATENTS, including, without limitation, any lapse,
       revocation, surrender, invalidation or abandonment of any of the PATENTS.

         Biogen may, in its sole discretion, decide to refrain from, or cease to
       prosecute, or maintain any of the PATENTS. In such an event, Biogen shall
       notify Tanox promptly and in sufficient time to permit Tanox at its sole
       discretion to continue such prosecution or maintenance at Tanox's
       expense. If Tanox elects to continue such prosecution or maintenance,
       Biogen shall execute such documents and perform such acts at Biogen's
       expense as may be reasonably necessary for Tanox to so continue such
       prosecution or maintenance.


                                       14
<PAGE>
5.2    In the event either party or its AFFILIATES, DISTRIBUTORS (or Tanox's
       SUBLICENSEES) becomes aware of any actual or probable infringement of a
       PATENT claim licensed to Tanox under this Agreement, it shall notify the
       other party in writing of the details to the extent known of such
       infringement. Tanox, in its sole discretion and at its sole expense, may
       take action against any alleged infringer but would be required to take
       such action in the name of Biogen, if legally permissible, and if Biogen
       consents thereto. In determining whether to bring an action to enforce
       any such PATENT, Tanox shall act in a commercially reasonable manner,
       giving due consideration to the threat represented by the infringement
       and the potential risk to the PATENT involved. In the event Tanox
       declines within six (6) months of notification of such infringement to
       either (i) cause infringement to cease such as, for example, by
       settlement, or (ii) initiate legal proceedings against the infringer.
       Biogen may, but is not obligated to (upon notice to Tanox) initiate legal
       proceedings against the infringer, at Biogen's expense and in its own
       name. Biogen, under the circumstances of the previous sentence, is not
       obligated to initiate proceedings against more than one infringer at a
       time.

5.3    In the event either Tanox or Biogen shall initiate or carry out legal
       proceedings to enforce any of the PATENTS licensed under this Agreement
       against an alleged infringer, the party not initiating or carrying out
       such proceedings shall fully cooperate with, and supply all reasonable
       assistance requested by, the other party. Except as described hereunder,
       any party that institutes any suit to protect or enforce any such PATENTS
       shall have control of that suit and shall bear the reasonable expenses
       incurred by the non-initiating party in providing such assistance and
       cooperation as is requested pursuant to this Article.

5.4    Any recovery obtained by Tanox as the result of legal proceedings
       initiated and paid for by Tanox to enforce any of the PATENTS licensed
       under this Agreement against an


                                       15
<PAGE>
       alleged infringer, whether obtained by settlement or otherwise, shall
       (after reimbursement of all otherwise unreimbursed legal fees and
       expenses incurred by either Tanox or Biogen) be paid 100% to Tanox, and
       such recovery obtained by Tanox shall be treated as NET SALES so that
       Tanox shall pay a royalty to Biogen commensurate with the recovery as
       specified under Article 3.2. Any recovery obtained by Biogen as the
       result of legal proceedings initiated and paid for by Biogen to enforce
       such PATENTS against an alleged infringer, whether obtained by settlement
       or otherwise, shall (after reimbursement of all otherwise unreimbursed
       legal fees and expenses incurred by either Biogen or Tanox) be paid 100%
       to Biogen.

6. INDEMNIFICATION

6.1    Tanox and its AFFILIATES and its SUBLICENSEES assume all risk of damage
       or injury to persons or property arising out of the clinical testing,
       manufacture, use, marketing, promotion, distribution, or sale of LICENSED
       PRODUCT(S). Tanox shall hold harmless and indemnify Biogen, its officers,
       directors, agents, shareholders and employees (the "Biogen Indemnitees")
       from and against any and all liabilities, damages, losses, costs and
       expense incurred or imposed upon Biogen Indemnitees or any one of them in
       connection with any claims, suits, actions, demands, proceedings, causes
       of action or judgments resulting from arising out of: (i) the
       development, design, preclinical or clinical testing, manufacture, use,
       marketing, promotion, distribution or sale of the LICENSED PRODUCT(S) by
       Tanox or any of its AFFILIATES, SUBLICENSEES or DISTRIBUTOR(S) or any of
       their respective agents or employees; (ii) any other activities carried
       out by Tanox or any of its AFFILIATES, SUBLICENSEES or DISTRIBUTOR(S) or
       any of their respective agents or employees, including failure to comply
       with applicable law.

   Biogen shall give prompt notice to Tanox of any claim that may be subject to
indemnification upon Biogen's receipt of notice to such claim, and Tanox shall


                                       16
<PAGE>
assume the defense thereof, including the employment of counsel reasonably
satisfactory to Biogen, provided, however, that Tanox shall act reasonably and
in good faith with respect to all matters related to settlement or disposition
of any claim as the settlement or disposition thereof relates to Biogen, and
further provided that Tanox shall not settle or otherwise dispose of any claim
without prior written notice to Biogen. Biogen shall have the right to employ
separate counsel in any such action and to participate in the defense thereof,
but the fees and expenses in this situation shall be Biogen's.

6.2    Following the FIRST COMMERCIAL SALE, Tanox shall purchase and maintain in
       effect, and require its AFFILIATES and SUBLICENSES to purchase and
       maintain in effect, a policy of product liability insurance in the amount
       of at least $10 million dollars (combined single limit) covering all
       claims with respect to any LICENSED PRODUCT used, made, sold, imported,
       licensed or otherwise distributed by Tanox or any of its AFFILIATES,
       DISTRIBUTOR(S) or SUBLICENSES within the term of this Agreement. Each
       policy obtained under this Article shall specify Biogen as an additional
       insured and Tanox shall furnish to Biogen upon Biogen's request, a
       certificate evidencing such insurance.

7.     REPRESENTATIONS, WARRANTIES AND LIABILITY

7.1    BIOGEN warrants that it owns title to the PATENTS listed in Appendix A
       and the TECHNOLOGY and has the right to enter into this Agreement. BIOGEN
       MAKES NO REPRESENTATION OR WARRANTY AS TO THE VALIDITY OF THE PATENTS.

7.2    BIOGEN MAKES NO REPRESENTATION OR WARRANTY THAT THE MANUFACTURE, USE,
       IMPORTATION OR SALE OF LICENSED PRODUCTS BY TANOX OR ITS SUBLICENSEES OR
       THEIR CUSTOMERS WILL NOT CONSTITUTE AN INFRINGEMENT OF THE INTELLECTUAL
       PROPERTY RIGHTS OF OTHERS.


                                       17
<PAGE>
7.3    BIOGEN MAKES NO REPRESENTATIONS, EXTENDS NO WARRANTIES OF ANY KIND,
       EITHER EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE IMPLIED
       WARRANTIES OF MERCHANT ABILITY OR FITNESS FOR A PARTICULAR PURPOSE, AND
       ASSUMES NO RESPONSIBILITY WHATEVER WITH RESPECT TO DESIGN, DEVELOPMENT,
       MANUFACTURE, USE, SALE, IMPORTATION OR OTHER DISPOSITION OF LICENSED
       PRODUCT BY TANOX OR ITS AFFILIATES, SUBLICENSEES OR DISTRIBUTOR(S) TO
       THEIR RESPECTIVE CUSTOMERS.

7.4    The entire risk as to performance of LICENSED PRODUCTS is assumed by
       Tanox and its AFFILIATES, DISTRIBUTOR(S) and SUBLICENSEES. Except as set
       forth under Article 6, in no event shall either party be responsible or
       liable to the other, its AFFILIATES or DISTRIBUTOR(S) (or Tanox's
       SUBLICENSEES), end users (or any other individual or entity regardless of
       legal theory), for any DIRECT, INDIRECT, SPECIAL, INCIDENTAL OR
       CONSEQUENTIAL DAMAGES OR LOST REVENUES OR PROFITS arising out of this
       Agreement. The provisions of this Article 7.4 shall apply even though a
       party may have been advised of the possibility of such damage.

7.5    Tanox covenants and agrees that in conducting activities contemplated
       under this Agreement, it shall comply with all applicable laws and
       regulations including those related to the manufacture, use, labeling,
       importation and marketing of LICENSED PRODUCT.


                                       18

<PAGE>
7.6    Tanox shall make no statements, representations or warranties or accept
       any liabilities or responsibilities to, or with regard to any person or
       entity, which are inconsistent with any disclaimer or provisions of this
       Article 7 and Tanox shall take reasonable steps to ensure that its
       AFFILIATES and its SUBLICENSEES do not do so.

7.7    Tanox and Biogen agree that, except for the obligations contained within
       this Agreement, there are no other express and/or implied obligations on
       either party that can be created by the activities contemplated
       hereunder.

8.     CONFIDENTIALITY

8.1    During the term of this Agreement, and for a period of seven (7) years
       after termination, Biogen and Tanox shall treat all confidential
       information (including confidential TECHNOLOGY) received from the other
       hereunder as the exclusive property of the disclosing party and each
       party agrees not to use or disclose to any third party any such
       information, except as permitted hereunder, without first obtaining the
       disclosing party's written consent. Each party further agrees to take all
       practicable steps to ensure that any such information shall not be used
       by its directors, officers, employees, or agents and that it shall be
       kept fully private and confidential by them.

8.2    The above provision of confidentiality shall not apply to that part of
       such information which a receiving party is clearly able to demonstrate:

   (a)   was fully in its possession prior to receipt from the other; or

   (b)   was in the public domain at the time of receipt from the other; or

   (c)   became part of the public domain through no fault of the party
         receiving such information, its director, officers or employees;

   (d)   was lawfully received without obligations of confidentiality or non-use
         from some third party having a right of further disclosure; or

   (e)   other than information submitted pursuant to obtaining regulatory
         approval for LICENSED PRODUCT, is required to be disclosed by law or
         applicable government or European Community regulations; provided,
         however, that the disclosing party is given prior written notice of
         such required disclosure and afforded an opportunity to participate in
         drafting a protective order or otherwise limiting the disclosure to the
         extent possible.


                                       19
<PAGE>
9.     COMMERCIALIZATION

9.1    Tanox undertakes to use reasonable commercial efforts to diligently
       develop and market LICENSED PRODUCT in the major market countries for the
       LICENSED PRODUCT in the TERRITORY and, with Biogen's reasonable
       assistance, to diligently obtain regulatory approval for LICENSED PRODUCT
       in the major market countries for the LICENSED PRODUCT in the TERRITORY.
       Tanox shall provide Biogen with summaries of such efforts, including the
       clinical development status of LICENSED PRODUCTS in the FIELD, before
       January 1 of each calendar year and shall provide Biogen with notice of
       the date Tanox and/or its SUBLICENSEES makes the FIRST COMMERCIAL SALE of
       LICENSED PRODUCT. For the purposes of this Article 9.1, "reasonable
       commercial efforts" means the usual practice followed by a
       biopharmaceutical company in pursuing commercialization of its products.

9.2    Tanox shall take reasonable steps to ensure the compliance of its
       SUBLICENSEES with all applicable laws and regulations, including, without
       limitation any labeling requirements relating to the sale of LICENSED
       PRODUCT by the SUBLICENSEES.

10.   PUBLICITY


                                       20
<PAGE>
10.1   Neither Tanox nor its AFFILIATES nor its SUBLICENSEES shall make any use
       of the name of Biogen in connection with the exercise of its rights
       hereunder (including in any advertising, promotional or sales
       literature), without the prior written consent of Biogen, except as
       required by law or regulation. Upon request, Tanox or its AFFILIATES or
       its SUBLICENSEES shall ensure that LICENSED PRODUCT will be labeled "sold
       under license" from Biogen.

10.2   Any initial announcements or similar publicity with respect to this
       Agreement shall be at such time and in such manner and such form as
       Biogen and Tanox shall mutually agree. Thereafter, either party may
       subsequently publicize the terms and subject matter of this Agreement in
       its sole discretion as long as the content of subsequent disclosures is
       consistent with the approved form. To the extent that any such
       publication or the terms and subject matter of this Agreement is
       inconsistent with the agreed announcement AND/or includes additional
       disclosure relating thereto, the party submitting any such subsequent
       announcement or similar publicity shall first send it to the other party
       for review. The other party agrees to review and return such announcement
       or similar publicity to the sending party within 24 hours of receipt.

11.    PATENT EXTENSIONS

11.1   Subject to the applicable governmental laws and regulations in the
       TERRITORY regarding extension of patent terms, Tanox and its SUBLICENSEES
       shall cooperate fully with Biogen in providing Biogen, at Biogen's
       request, all facts and documentation which may assist Biogen in its
       procurement of term extension for the PATENTS. Biogen shall be
       responsible for, and shall bear the expense, of obtaining such patent
       term extension as to patent rights encompassing LICENSED PRODUCT.


                                       21
<PAGE>
11.2   Upon termination of this Agreement by Biogen pursuant to Article 12.2,
       information related to government approvals for LICENSED PRODUCT and
       safety, efficacy, and toxicity studies of the LICENSED PRODUCT (provided
       that the above is accessible to Tanox and/or its SUBLICENSEES in the
       TERRITORY) shall be shared by Tanox and/or its SUBLICENSEES with Biogen
       and such information may be transmitted to a governmental agency for use
       by Biogen if necessary.

12.    TERM AND TERMINATION

12.1   This Agreement shall commence on the Effective Date and shall continue
       until Tanox's obligation to pay royalties pursuant to Article 3
       terminates.

12.2   If the parties agree that this Agreement has been breached as to a
       material covenant, undertaking, representation or obligation and that the
       allegedly breaching party has not pursued steps to correct or cure such
       breach within sixty (60) days of notification from the other party, or if
       a court of competent jurisdiction so determines in a decision that is
       unappealable or unappealed within the time allowed for appeal, or if the
       decision is affirmed on appeal, by the highest court with jurisdiction,
       then the non-breaching party shall have the right, by notice in writing,
       to terminate this Agreement. A party shall also have the right to
       terminate this Agreement in the event that the other party shall enter
       into any arrangement or composition with its creditors, or enter or be
       put into voluntary or compulsory liquidation or bankruptcy (except for
       the purpose of any reorganization reasonably acceptable to the other
       party), or have its business enjoined into receivership by executive or
       judicial authorities.

12.3   Tanox shall have the right to terminate this Agreement on thirty (30)
       days written notice to Biogen, provided that Tanox supplies to Biogen
       within 180 days of termination hereunder, and sooner if feasible, all
       preclinical, clinical and other data reasonably related to Biogen's


                                       22
<PAGE>
       relicensing of the LICENSED PRODUCT as well as a royalty-free,
       nonexclusive license to any Tanox-owned patents necessary to make, use,
       or sell the LICENSED PRODUCT. Any termination under this Article 12 shall
       be without prejudice to the rights of either party against the other then
       accruing or otherwise accrued under the Agreement.

12.4   Expiration of this Agreement pursuant to Article 12.1 shall result in the
       exclusive license to Tanox under Article 2.1 being converted to a
       non-exclusive and cost and royalty-free license. Both expiration of the
       Agreement under Article 12.1 and termination of this Agreement by either
       party pursuant to Articles 12.2 and 12.3 shall terminate all outstanding
       obligations and liabilities between Biogen and Tanox arising from this
       Agreement except:

       (a)  obligations to pay royalties and other sums accruing hereunder up to
            the day of such termination or expiration;

       (b)  obligations for record keeping and accounting reports for so long as
            LICENSED PRODUCT is sold pursuant to this Agreement up to the date
            of termination or expiration. Tanox shall render a final report
            along with any royalty payment at such time after termination or
            expiration of this Agreement;

       (c)  Biogen's right to inspect books and records up to the date of
            termination or expiration as in Article 3.7;

       (d)  obligations of defense and indemnity under Article 6;

       (e)  any cause of action or claim of Tanox or Biogen accrued as of the
            date of termination or expiration because of any breach or default
            by the other party hereunder;

       (f)  the confidentiality provisions of Article 8;

       (g)  the disclosure obligations of Article 11.2, as applicable.

       (h)  all other terms, provisions, representations, rights and obligations
            contained in this Agreement that by their sense and context are
            intended to survive until performance thereof by either or both
            parties; and

       (i)  the right to complete the manufacture and sale of LICENSED PRODUCTS
            which qualify as "work in progress" under generally accepted cost
            accounting standards or which are in stock at the date of
            termination, and the obligation to pay royalties on NET SALES of
            such LICENSED PRODUCTS.


                                       23
<PAGE>
13.    NOTICES

13.1   Any notice required or permitted to be given hereunder shall be sent in
       writing by registered or certified airmail, postage prepaid, return
       receipt requested, or by telecopier, air courier or hand delivery,
       addressed to the party to whom it is to be given as follows:

   If to BIOGEN:   Biogen, Inc.
                   14 Cambridge Center
                   Cambridge, MA 02142
                   Telephone (617) 679-2000; Fax (617) 679-2838
                   Attention: Vice President-General Counsel
   If to TANOX:
                   Tanox Biosystems, Inc.
                   10301 Stella Link
                   Houston, TX 77025
                   Telephone: (713) 664-2288; Fax: (713) 664-8914
                   Attention: Nancy T. Chang, Ph.D., President
                   and CEO

   or to such other address or addresses as may from time to time be given in
   writing by either party to the other pursuant to the terms hereof.


                                       24
<PAGE>
13.2   Any notice sent pursuant to this Article shall be deemed delivered within
       five (5) days if sent by registered or certified airmail and within
       twenty-four (24) hours if sent by telecopier, air courier or hand
       delivery.

14.    EXPORT LAWS AND REGULATIONS OF THE UNITED STATES
      The Export regulations of the United States Department of Commerce
prohibit the exportation from the United States of certain types of technical
data and commodities unless the exporter (i.e., Tanox, AFFILIATES,
DISTRIBUTOR(S) or the SUBLICENSEES) has received the required license. In
addition, the exporter may be required to obtain certain written assurances
regarding re-export from the foreign importer for certain types of technical
data and commodities. Tanox agrees to comply with (and shall take reasonable
steps to ensure the compliance of its AFFILIATES, DISTRIBUTOR(S) and its
SUBLICENSEES with) the Export Administration Regulations of the United States
Department of Commerce.

15.    MISCELLANEOUS

15.1   ENTIRE AGREEMENT:

       This Agreement constitutes the entire understanding between the parties
with respect to the subject matter between the parties with respect to the
subject matter hereof, and supersedes and replaces all prior agreements,
understandings, writings and discussions between parties relating to said
subject matter.

15.2  AMENDMENTS: WAIVERS:
      This Agreement may be amended and any of its terms or conditions may be
waived only by a written instrument executed by both parties, or, in the case of
a waiver, by the party waiving compliance. The failure of either party at any
time to require performance of any provision hereof shall in no manner affect
its rights a later time to enforce the same. No waiver by either party of any
condition or term in any instance shall be construed as a further or continuing
waiver of such condition or term or of another condition or term.

                                       25
<PAGE>
15.3  NO AGENCY:
      The relationship between Tanox and Biogen is that of independent
contractor. Nothing herein shall be deemed to constitute Tanox, on the one hand,
or Biogen, on the other hand, as the agent or representative of the other, or as
master and servant, employer and employee, joint venturers or partners for any
purpose.

15.4  ASSIGNMENT:
      This Agreement shall not be assigned by Tanox without the prior written
consent of Biogen, except to an AFFILIATE or a successor Tanox's entire
business. This Agreement shall be binding upon and inure to the benefit of and
be enforceable by the parties hereto and their respective successors and
permitted assigns.

15.5  LAW OF THE CONTRACT:
      This Agreement shall be governed by and construed and interpreted in
accordance with the law of the Commonwealth of Massachusetts.

15.6  SEVERABILITY:
      In the event one or more provisions of this Agreement should for any
reason be held by any court or authority having applicable jurisdiction to be
invalid, illegal or unenforceable, such provision(s) shall either be reformed to
comply with applicable law or stricken if not so conformable, so as not to
affect the validity or enforceability of the remainder of this Agreement.

                                       26
<PAGE>
15.7  AGREEMENT TO PERFORM NECESSARY ACTS:
      Each party agrees to perform further acts and execute and deliver any and
all further documents, agreements, and/or instruments which may be reasonable to
carry out or effect the provisions of this Agreement.

15.8  COUNTERPARTS:
      This Agreement may be executed in counterparts, and each such counterpart
shall be deemed an original for all purposes.

   IN WITNESS WHEREOF, the parties hereto have caused this Agreement to be
executed by their duly authorized officers on the date and year first above
written.



BIOGEN, INC.                              TANOX BIOSYSTEMS, INC.

By: /s/ MICHAEL J. ASTRUE                 By: /s/ NANCY T. CHANG, Ph.D.

Name: Michael J. Astrue                   Name: Nancy T. Chang, Ph.D.

Title: Vice President-General Counsel     Title: President and CEO

Date: June 18, 1998                       Date: June 23, 1998



                                       27
<PAGE>
                                   APPENDIX A


             
COUNTRY          NUMBER          FILED       ISSUED #      EXPIRES       BIOGEN NO.

Australia          *            11/27/91      662891      11/27/011          *

Canada             *            11/27/91        --           --              *

EPO*            92903295.1      11/27/91      512112       11/27/011         *

Japan              *            11/27/91        --           --              *

United          07/916,098      11/27/91        --                           *
States           to issue
Q2  1998



   *Austria, Belgium, Switzerland, Germany, Denmark, Spain, France, Great
Britain, Greece, Italy, Liechtenstein, Luxembourg, Netherlands, Sweden


                                       28
<PAGE>
                                   APPENDIX B

   The following are patents and patent applications defined as PROTEIN DESIGN
LABS PATENTS and shall expressly include any United States continuations,
continuations-in-part or divisions thereof or any substitute applications
therefor; or foreign counterparts thereof, any patents issued with respect to
such patent applications, any reexaminations, reissues, extensions or patent
term extensions of any such patents.

   1. United States Patent Numbers 5,585,089, 5,693,761, 5,693,762, and U.S.
patent divisional application numbers 08/477,728, 08/474,040 and 08/487,200 of
issued United States Patent No. 5,530,101.

   2. European Patent 0451216

   3. Japanese Patent application No. 4-503758


                                       29
<PAGE>
                                   APPENDIX C

                       TECHNOLOGY TO BE PROVIDED BY BIOGEN

      1. Purified murine antibody 5A8 (10mg) and its complete characterization
information, including:

         a. effect of 5A8 on immune functions;

         b. effect of 5A8 on apoptosis of CD4 T lymphocytes; and

         c. neutralization data on HIV-1 primary isolates.

      2. Purified humanized 5A8 (10mg) and its complete characterization
information, including:

         a. molecular design of the humanized version;

         b. procedures of testing and selecting the humanized version; and

         c. neutralization data on HIV-1 primary isolates.

      3. Purified humanized or human IgG4 irrelevant control (10mg).

      4. Purified recombinant soluble CD4 (5mg).

      5. Purified recombinant HIV-1 gp120 (5mg).

      6. Sera from monkeys immunized with humanized 5A8 for anti-id response
studies.

      7. Cell lines:

         a. murine hybridoma producing 5A8; and

         b. production cell line for humanized 5A8.

                                       30
<PAGE>
      8. Plasmids containing the antibody genes of humanized 5A8 and complete
information on their construction.

      9. Production medium for humanized 5A8 (powder, 20 liters).

      10. Preclinical data:

         a. data on the analysis of anti-id response in monkeys immunized with
humanized 5A8;

         b. complete reports for all toxicology studies (monkeys and rodents)
and any other pre-clinical safety studies performed;

         c. complete report for human tissue cross-reactivity; and

         d. complete reports of all pharmacology studies.

      11. Process descriptions and manufacturing batch records:

         a. cell blank preparation and characterization;

         b. cell line testing reports;

         c. cell culture process (culture medium formulation and preparation,
and cell growth conditions);

         d. recovery downstream process (buffer compositions and purification
methodologies);

         e. formulation of the final product and stability studies;

         f. filling and vialing; and

         g. release testing specifications and testing protocols.

                                       31
<PAGE>
FAX TRANSMISSION

To:    Dr. Linda Burkly
       Biogen, Inc.

Date:  December 11, 1996

From:  Keith A. Reimann
       Division of Viral Pathogenesis
       Department of Medicine
       Beth Israel Hospital - RE-113
       330 Brookline Avenue
       Boston, MA 02215

       Phone-       617-667-4583
       Fax          617-667-8210
       E-mail       kreimann@bih.harvard.edu

Re:    Hu5A8 IN VIVO summary

Attached is the experimental protocol and two figures showing data to date. In
one monkey we get an 80% reduction in viral load; in the other >95% reduction.
In the monkey with the greatest reduction in viral load, we saw a substantial
increase in CD4 count (don't know what this means since we ALWAYS see CD4
increases....even in uninfecteds).

With AZT monotherapy there is usually less than 1 log decrease in HIV RNA. With
protease inhibitor monotherapy there is about 1-2 log decrease. Based on these
data, I think we're in the ball park of efficacy as compared with conventional
anti-retrovirals. We have additional specimens as the indicated time points that
have NOT been analyzed for RNA.

Let me know what you think.

Keith

<PAGE>
HUMANIZED 5A8 (HU5A8) THERAPEUTIC EFFECT, EXPERIMENT 3.

EXPERIMENTAL PROTOCOL

OBJECTIVE

These studies are a pilot experiment to assess the antiviral activity of hu5A8
in a chronic SIVmac infection model. Two SIVmac-infected rhesus monkeys will be
treated four times with hu5A8 (3 mg/kg) I.V. over 10 days and changes in plasma
viral RNA levels will be assessed.

ANTIBODY

Humanized 5A8.3 - Lot 2927 1.20.1993
6.6 mg/ml

ANIMAL PAIR 2:

Mm 74-84 (SIVmac infected on May 28, 1996)
Mm 67-85 (SIVmac infected on May 28, 1996)

Sampling time points:

         TIME                         DATE        PLASMA        CBC/PHENOTYPING
         -----------------------    ---------    ---------    ------------------
         Day  -8                     04 Nov          X                 X
   o     Day  -4                     08 Nov          X                 X

   o     Day  0; pre                 12 Nov          X                 X
         Day  0; 10-15 min post      12 Nov          X                 X

   o     Day  2; pre                 14 Nov          X                 X
         Day  2; 10-15 min post      14 Nov          X                 X

   o     Day  6; pre                 18 Nov          X                 X
         Day  6; 10-15 min post      18 Nov          X                 X

         Day 10; pre                 22 Nov          X                 X
         Day 10; 10-15 min post      22 Nov          X                 X

   o     Day 13                      25 Nov          X                 X
         Day 20                      02 Dec          X                 X

         Day 27                      09 Dec          X                 X
         Day 41                      23 Dec          X                 X
<PAGE>
HUMANIZED 5A8 (HU5A8) THERAPEUTIC EFFECT, EXPERIMENT 3.

TREATMENT NOTES

Day  -8            04 Nov  Weights:
                           Mm 174-84  10.6 kg
                           Mm 167-85  10.6 kg

Day  -4            08 Nov  Dosage calculation:  10.6 kg x 3mg/kg = 31.8 mg
                                                 31.8mg/6.6mg/mi = 4.8 mi
                           Mm 174-84  10.6 kg
                           Mm 167-85  10.5 kg

Day  0             12 Nov  Antibody given

Day  2             14 Nov  Antibody given

Day  6             18 Nov  Antibody given

Day 10             22 Nov  Antibody given
<PAGE>
                                     [INSERT GRAPH HERE]

                                     [INSERT GRAPH HERE]

FIG 11 - Hu5A8 dramatically decreases viral RNA and increases CD4 cells in
SIVmac-infected monkeys. Two rhesus monkeys received 3 mg/kg injections of hu5A8
as indicated by arrows. CD4 cells in blood were measured (A) and viral RNA was
quantitated by bDNA assay (B).
<PAGE>
HUMANIZED 5A8 (hu5A8) THERAPEUTIC EFFECT, EXPERIMENT 3.

SPECIMENS COLLECTED/ANALYSES PERFORMED

Plasma --      EDTA plasma will be collected within 4 hours of draw. One aliquot
               of 1 mL (accurately measured) for viral RNA measurement will be
               stored at -70 C. Any remaining plasma will be stored at -70 C as
               well.

               It is important that plasma be completely separated from cells
               and that it is quickly stored at -70 C. Thawing and refreezing
               should be avoided.

CBC --         A complete blood count with automated partial differential will
               be performed on a Coulter T540 analyzer.

Phenotype --   PBL will be phenotyped using the following antibody combinations:

                  anti-human IgG4-FITC (coating)*
                  CD3/CD4
                  CD3/CD8

                   *anti-human IgG4 will be used starting on Day 0.

SIV RNA --     Viral RNA is plasma will be measured by Chiron Diagnostics using
               a branched-chain DNA assay modified to detect SIV gag RNA.

<PAGE>
AIDS RESEARCH AND HUMAN RETROVIRUSES
Volume 13, Number 11, 1997
Mary Ann Liebert, Inc.

A HUMANIZED FORM OF A CD4-SPECIFIC MONOCLONAL ANTIBODY EXHIBITS
DECREASED ANTIGENICITY AND PROLONGED PLASMA HALF-LIFE IN RHESUS MONKEYS
WHILE RETAINING ITS UNIQUE BIOLOGICAL AND ANTIVIRAL PROPERTIES

KEITH A. REIMANN(1), WENYU LIN(1), SARAH BIXLER(2), BETH BROWNING(2), BARBARA N.
EHRENFELS(2), JODIE LUCCI(2), KONRAD MIATKOWSKI(2), DIAN OLSON(2), THOMAS H.
PARISH(2), MARGARET D. ROSA(2), FREDERICK B. OLESON(2), YEN MING HSU(2), EDUARDO
A. PADLAN(3), NORMAN L. LETVIN(1), and LINDA C. BURKLY(2)


ABSTRACT

Certain monoclonal antibodies (MAbs) directed against CD4 can efficiently block
HIV-1 replication IN VITRO. To explore CD4-directed passive immunotherapy for
prevention or treatment of AIDS virus infection, we previously examined the
biological activity of a nondepleting CD4- specific murine MAb, mu5A8. This MAb,
specific for domain 2 of CD4, blocks HIV-1 replication at a post-gp120-CD4
binding step. When administered to normal rhesus monkeys, all CD4+ target cells
were coated with antibody, yet no cell clearance or measurable immunosuppression
occurred. However, strong anti-mouse Ig responses rapidly developed in all
monkeys. In the present study, we report a successfully humanized form of mu5A8
(hu5A8) that retains binding to both human and monkey CD4 and anti-AIDS virus
activity. When administered intravenously to normal rhesus monkeys, hu5A8 bound
to all target CD4+ cells without depletion and showed a significantly longer
plasma half-life than mu5A8. Nevertheless, an anti-hu5A8 response directed
predominantly against V region determinants did eventually appear within 2 to 4
weeks in most animals. However, when hu5A8 was administered to rhesus monkeys
chronically infected with the simian immunodeficiency virus of macaques,
anti-hu5A8 antibodies were not detected. Repeated administration of hu5A8 in
these animals resulted in sustained plasma levels and CD4+ cell coating with
humanized antibody for 6 weeks. These studies demonstrate the feasibility of
chronic administration of CD4-specific MAb as a potential means of treating or
preventing HIV-1 infection.

                                  INTRODUCTION

THE CD4 MOLECULE IS a high-affinity receptor for the envelope glycoprotein of
human immunodeficiency virus type 1 (HIV-1) and is, therefore, a potential
target for the prevention or treatment of AIDS(1). Certain CD4-specific
monoclonal antibodies (MAbs) can effectively block HIV-1-induced syncytium
formation and infection of lymphocytes and macrophages IN VITRO with extreme
efficiency(2-4), suggesting that CD4-directed MAb may have utility as an AIDS
therapy. We have studied extensively one such MAb, mu5A8, that binds similarly
to human and rhesus monkey CD4(5). This antibody maps to domain 2 of CD4 and
efficiently inhibits HIV-1 replication and virus- induced cell fusion acting at
a post-CD4-gp120 binding step(6,7). The IN VITRO antiviral activity of

<PAGE>
mu5A8 is similar when tested against the related virus, simian immunodeficiency
virus of macaques (SIVmac)(5).

   Previously we have shown that this MAb binds to all target CD4+ lymphocytes
without inducing cell clearance when administered intravenously to rhesus
monkeys.5 In fact, there is a transient increase in circulating CD4+ cells
following administration of the MAb. However, in normal rhesus monkeys, a strong
anti-mouse immunoglobulin response developed soon after administration that
blocked the mu5A8-CD4 binding, induced rapid plasma clearance, and, thus,
precluded its further use.

   In an effort to reduce the antigenicity of this MAb, we have developed a
"humanized" form of the mu5A8 MAb (hu5A8). Here we show that hu5A8 retains the
biological activity of the murine MAb, including similar binding to human and
nonhuman primate CD4 and potent antiviral activity against HIV-1 and SIVmac.
This humanized antibody, however, has a significantly longer plasma half-life
than the murine form and plasma levels could be maintained in normal rhesus
monkeys for more than 2 weeks. Although the immune response against hu5A8 was
reduced, a response eventually developed that was directed entirely against the
murine V region determinants. Interestingly, anti-hu5A8 responses were not
detected when hu5A8 was administered for 6 weeks to monkeys chronically infected
with the immunosuppressive virus SIVmac. These studies indicate that alteration
of MAbs to decrease their antigenicity can positively affect the
pharmacokinetics without altering the biological activity.

                              MATERIALS AND METHODS

CONSTRUCTION AND EXPRESSION OF hu5A8

   CLONING OF MURINE 5A8 HEAVY AND LIGHT CHAIN VARIABLE REGIONS. The mu5A8 heavy
chain variable (VH) region was amplified from genomic DNA of the murine
hybridoma cell line by polymerase chain reaction (PCR) using primers identical
in sequence to VH1BACK and VH1FOR as previously described(8), except that the
primer corresponding to VH1FOR lacked the two 3' nucleotides. The PCR
product was subcloned by blunt end ligation into a pUC vector and designated
pMDR904, and encodes amino acids 2-122 (AA(2)-AA(122)) of the 5A8 VH region. The
numbering system for immunoglobulin amino acids used throughout is that of Kabat
ET AL.(9) By comparison to known complementarity-determining regions (CDRs) and
framework (FR) sequences, the CDRs for the mu5A8 VH region are identified as
follows: CDR1 is AA(31)-AA(35), CDR2 is AA(50)-AA(66), and CDR3 is
AA(99)-AA(111). The mu5A8 light chain variable (V1) region was PCR amplified
from cDNA using primers ACE149 and ACE150, with cDNA first prepared from
poly(A)+ RNA of the murine hybridoma line by reverse transcription (RT) with the
same primers. The PCR fragment encoding AA1-AA(111) of the mu5A8 VL region was
subcloned by blunt end ligation into a pUC expression vector and designated
pMDR927. Mu5A8 VL CDRs are as follows: CDR1 is AA(24)-AA(40). CRD2 is
AA(56)-AA(62), and CDR3 is AA(95)-AA(102).

   HUMANIZATION OF THE 5A8 MONOCLONAL ANTIBODY. The amino acid sequences of
mu5A8 VH and VL regions were compared to the known human immunoglobulin
sequences to find the human FR sequences most closely matching the mu5A8 FR
sequences. Humanized versions were then

<PAGE>
designed by grafting the CDRs of the mu5A8 VH and VL sequences onto these human
FRs, MO30 and VJI, respectively.10,11 A small number of residues in the human
FR, which we believe will probably be critical to the three-dimensional
structure of the antibody combining site,12 were altered in order to match the
corresponding sequence of the mu5A8 FR. These hu5A8 VH and hu5A8 VL sequences
were then fused to genomic DNA encoding the human IgG4 constant region and the K
constant (CK) region, respectively. The 5A8 VH and VL regions were humanized by
synthesizing a series of oligonucleotides that would span the entire CDR-grafted
VH and VL sequences when ligated together. Eleven oligonucleotides (oligos) were
kinased, combined, and ligated to construct the VH region (oligos 312-120 to
312-131) and 12 oligos (312-132 to 312-143) were used to construct the VL region
with restriction sites contained in the flanking oligos to facilitate
subcloning. For the VH region, ligated oligos were digested with PSTI and BSTE
II and a 326-bp fragment subcloned into pLCB7, thereby substituting hu5A8 VH for
mu5A8 VH sequence, fusing the hu5A8 VH (at the PSTI site) 3' of a heavy chain
signal sequence and (at the BSTE II site) 5' of a splice site, generating
pMDR1032 and used to construct an hu5A8 VH expression vector. For the VL region,
ligated oligos were digested with ECORV and BGL II and a 330-bp fragment
isolated and subcloned into a pUC vector, generating pBAG176. After the sequence
was confirmed, the ECOR V-BGL II fragment encoding hu5A8 VL (AA1-AA112) was
isolated and used to construct an hu5A8 VL expression vector. Figure 1 shows the
deduced amino acid sequences from the hu5A8 VH and VL regions aligned with the
mu5A8 VH and VL sequences and the human antibody sequences with framework
regions that most closely match the murine sequences. The framework sequences of
hu5A8 VH and VL retained 12 and 5 murine-derived residues, respectively.

   CONSTRUCTION OF GLUTAMINE SYNTHETASE EXPRESSION VECTOR FOR THE hu5A8
MONOCLONAL ANTIBODY. A single expression vector containing separate
transcription units for the hu5A8 MAb heavy and light chains and carrying the
glutamine synthetase (GS) selection marker was constructed using the pEE6 and
pEE12 vectors developed and provided by CellTech Limited (Slough, UK).13 The
pEE6 and pEE12 vectors were each modified by insertion of a novel NOTI site into
the SMAI cloning site with NOTI linker pGCCGGCACT (New England Biolabs, Beverly,
MA) and designated pEE6.NOTI and pEE12.NOTI, respectively.

   The hu5A8 light-chain transcription unit was constructed by inserting a
1693-bp NOTI fragment encoding a K signal sequence fused to the hu5A8 VL and CK
regions into NOTI-linearized pEE12.NOTI, generating pMDR1042. This 1693-bp NOTI
insert was derived from pBAG177, which had been previously constructed by fusing
DNA containing the human K signal sequence (a 572-bp AatII-ECOR V fragment), the
hu5A8 VL (ECOR V-BGLII fragment of pBAG176), and the CK region (1276-bp
BCLI-AATII fragment) in 5' to 3' order, with NOTI sites flanking the K signal
and CK sequences. The K signal and CK sequences were in turn derived as follows.
A prototype signal sequence for the hu5A8 VL expression vector was constructed
by ligating oligos 360-81 and 360-82 followed by subcloning into a pUC vector,
generating pMDR985. AATII-ECOR V digestion of pMDR985 then released a DNA
fragment containing the K signal sequence. The CK region was cloned by PCR
amplification of a 2.5-kb ECORI fragment from human placental genomic DNA with
primers 370-54 and 370-55, with VK AA112 (lysine) and a splice site constructed
at the 5' end. The 1240-bp PCR fragment was subcloned into a pUC vector,
generating pSAB153, and further PCR amplified with primers 360-83 and 370-55 in
order to create a BCLI site at the 5' end upstream of VK AA112. The resultant
1276-bp PCR fragment was subcloned into a pUC vector (generating

<PAGE>
pMDR986) and DNA containing the CK region (AA108-AA214 according to Kabat amino
acid numbering) released by BCLI-AATII digestion.

   The hu5A8 heavy chain transcription unit was constructed in pEE6.NOTI by
ligating a NOTI- HINDIII fragment from pMDR1032 (encoding the H chain signal
sequence, hu5A8 VH region, and splice site) and a HINDIII-NOTI fragment encoding
the huIgG4 constant region together with NOTI- linearized pEE6.NOTI, thereby
generating pMDR1035. The human IgG4 constant region was previously cloned by PCR
amplification from human genomic placental DNA with primers 370-38 and 370-40,
the 2109-bp PCR fragment cloned by blunt ligation into a pUC vector (designated
pBAG101), and this construct further modified (pMDR1033) to remove BGLII and
BAMHI sites in the constant region sequence contained on a HINdIII-NOTI
fragment. To generate a GS vector carrying transcription units for both the
hu5A8 heavy and light chains, the BAMHI-BGLII fragment containing the H chain
transcription unit from pMDR1035 was inserted into the BAMHI site of pEE12.NOTI
carrying the hu5A8 L chain cassette, finally constructing pMDR1045.

     OLIGONUCLEOTIDES. Primers used for PCR amplification of the mu5A8 VH DNA
sequences were VHIBACK and VHIFOR, and for VL DNA sequences were ACE 149 and ACE
150. See Fig. 2 for sequences of primers used in this study.

   GENERATION OF HU5A8-PRODUCING CELL LINE. Plasmid DNA pMDR1045 containing
hu5A8 H and L chain sequences in the GS expression vector was linearized with
SALI and introduced into NS/0 murine myeloma cells(14) provided to Biogen
(Cambridge, MA) by CellTech Limited using the electroporation protocol of
Bebbington ET AL(13). Cells were routinely maintained in Iscove's modified
Dulbecco's medium (IMDM) (Sigma Chemical Company, St. Louis, MO) supplemented
with 10% heat-inactivated fetal bovine serum and 4 mM L-glutamine. After
transfection, cells capable of growth in glutamine-free IMDM selection medium
were selected. Clones were screened for hu5A8 production with an enzyme-linked
immunosorbent assay (ELISA) similar to that described below, detecting hu5A8 mAb
levels binding to recombinant soluble CD4 (rsCD4)-coated plates. Minor
differences were that plates were blocked with phosphate-buffered saline-1%
bovine serum albumin (BSA)-0.05% Tween 20, and 5A8 MAbs were detected with
rabbit anti-human IgG Fc (Jackson ImmunoResearch, West Grove, PA) followed by
horseradish peroxidase (HRP)-conjugated goat anti-rabbit immunoglobulin
(Bio-Rad, Hercules, CA), with each of these steps carried out for 1 hr at room
temperature. Tetramethylbenzidine (TMB) was added as substrate (420 MM), the
reaction stopped after dark blue color developed with the highest concentration
of standard and plates read at 450 mm. hu5A8 purified from medium after
transient expression in Cos7 cells was used as a standard. Positive clones were
selected and expanded for production, and hu5A8 was purified from culture
supernatant by protein A affinity chromatography. hu5A8 production by NS/0
transfectants achieved levels of at least 50 mg/liter.

   mu5A8 was produced and purified as described previously(6). Human IgG4 (TSI
Center for Diagnostic Products, Milford, MA) was used as a control antibody for
IN VIVO studies.

ANIMALS AND ANTIBODY ADMINISTRATION

<PAGE>
   The rhesus monkeys (MACACA MULATTA) used in this study were maintained in
accordance with the guidelines of the Committee on Animals for the Harvard
Medical School and the GUIDE FOR THE CARE AND USE OF LABORATORY ANIMALS.15 In
some experiments, monkeys were previously infected by intravenous inoculation
with uncloned simian immunodeficiency virus of macaques (SIVmac), strain 251,
which had been propagated in human peripheral blood mononuclear cells (PBMCs).

   Monkeys were administered either mu5A8, hu5A8, or a control human IgG4
diluted to 1-2 mg/ml in PBS by intravenous injection over approximately 1 min.
In experiments where hu5A8 was administered chronically, monkeys received
intravenous injections every 7 days. Monkeys were anesthetized with ketamine
hydrochloride for all procedures.

ANTIBODY BINDING TO HUMAN AND MONKEY CD4

   The binding of mu5A8 and hu5A8 to human CD4 was compared using the
CD4-expressing cell line, Jurkat (American Type Culture Collection, Rockville,
MD). Cells were incubated for 30 min at 37(degree)C with mu5A8 or hu5A8 that had
been serially diluted in PBS-0.5% BSA. Cells were washed and then stained with
either goat anti-mouse IgG or goat anti-human IgG antisera conjugated to
fluorescein isothiocyanate (FITC) and relative fluorescence determined on a flow
cytometer. To assess binding of these MAbs to rhesus monkey CD4, peripheral
blood lymphocytes (PBLs) from normal rhesus monkeys were isolated by
Ficoll-ditrizoate density gradient centrifugation, incubated with serially
diluted hu5A8 or mu5A8 as described previously, and stained with goat anti-human
or anti-mouse immunoglobulin. To permit gating on CD4+ monkey PBLs, cells were
stained with noncompeting CD4 domain 3-specific antibody conjugated to
phycoerythrin (OKT4-PE; Ortho Diagnostic Systems, Raritan, NJ).(6) However,
prior to staining with OKT4-PE, cells were incubated with mouse immunoglobulin
to saturate any unoccupied goat anti-mouse immunoglobulin binding sites that
could potentially bind OKT4-PE. Relative binding of hu5A8 and mu5A8 on CD4+ PBLs
was assessed flow cytometrically using the phycoerythrin channel to gate on CD4+
lymphocytes and the fluorescein channel to determine relative binding of
antibody. Data were expressed as a percentage of the maximum fluorescence
intensity measured.

INHIBITION OF HIV-1-INDUCED SYNCYTIUM FORMATION

   The antibodies mu5A8 and hu5A8 were tested for their ability to block
syncytium formation between HTLV-IIIB-infected H9 cells and uninfected C8166
cells using the procedure described previously.(6,16) Results were expressed as
percentage inhibition relative to the maximum number of syncytia per well in the
absence of monoclonal antibody.

INHIBITION OF SIVMAC REPLICATION in vitro

   The ability of mu5A8 and hu5A8 to suppress SIVmac replication in monkey
lymphocytes was tested using an IN VITRO viral outgrowth assay similar to that
previously described.5 Briefly, lymphocytes isolated from peripheral blood of
rhesus monkey chronically infected with SIVmac were depleted of CD8+ cells using
monoclonal antibody and immunomagnetic beads. CD8-depleted cultures were
activated with concanavalin A (5 Mg/ml) and cultured for 2 weeks in the presence
of varying concentrations of mu5A8, hu5A8, or an isotype-matched murine antibody
of irrelevant

<PAGE>
specificity. Virus replication in these cultures was determined by measuring
SIVmac p27 core antigen in culture supernatants using a commercial kit (SIV p27
core antigen kit; Coulter Corp., Miami, FL).

SERUM MONOCLONAL ANTIBODY LEVELS

   mu5A8 MAb levels in serum were measured as previously described.5 Serum
levels of hu5A8 were determined by coating microtiter plates with human
recombinant soluble CD4 at 5 Mg/ml in 0.5 M bicarbonate buffer (pH 9.0), 50
Ml/well overnight at 4(degree)C. Plates were washed with PBS, blocked with
PBS-2% nonfat dry milk for 2 hr at room temperature, washed again, and monkey
serum diluted in PBS-2% milk was added at 50 Ml/well for 2 to 3 hr at room
temperature. After washing, hu5A8 MAb levels were detected with HRP-conjugated
anti-human IgG4 at a 1:5000 dilution, 50 Ml/well, for 1 hr at room temperature.
After final washes, O-phenyldiamine substrate (OPD; Calbiochem, La Jolla, CA)
was added at 100 Ml/well for 20 min, the reaction stopped with 1 N H2SO4 (100
Ml/well), and absorbance read at 490 nm on a microplate reader. Data are
expressed as micrograms per milliliter using purified hu5A8 as a standard.

IMMUNOPHENOTYPING OF MONKEY PERIPHERAL BLOOD LYMPHOCYTES

   The absolute number of CD4+ lymphocytes in peripheral blood was determined
using human CD4-specific monoclonal antibodies, whole-blood lysis technique, and
routine flow cytometric analysis as described for immunophenotyping rhesus
monkey leukocytes.(17) To determine whether circulating PBLs were coated with
hu5A8, cells were stained with FITC-conjugated sheep anti- human IgG4 (The
Binding Site, Birmingham, UK) and counterstained with a CD4 domain 3- specific
antibody as described previously.

ANTI-IMMUNOGLOBULIN LEVELS

   Microtiter plates were coated with hu5A8, human IgG4 specific for VLA4
(generously provided by R. Lobb, Biogen), or with mu5A8 to measure the humoral
response in monkeys directed against the hu5A8 MAb, as well as to determine the
specificity of the response against the human immunoglobulin constant region and
against the 5A8 idiotype. All proteins were used to coat plates at 5 Mg/ml in
0.5 M bicarbonate buffer (pH 9.0), 50 Ml/well overnight at 4(degree)C. Plates
were washed with PBS, blocked with PBS-2% milk, washed again, and monkey serum
added as described above. Anti-hu5A8 serum levels were detected with
HRP-conjugated anti-human IgG1 + IgG2 + IgG3 (Zymed Laboratories, South San
Francisco, CA) at 1:1000 dilution and OPD substrate then employed as described
above. Data are expressed as the reciprocal of the highest serum dilution that
gave a signal greater than 2 standard deviations over the signal obtained from
normal monkey serum.

   As an alternative means of determining the fine specificity of the
anti-immunoglobulin response, we employed the method of Cobbold, ET AL.(18)
Briefly, hu5A8 (1 Mg/ml) was immobilized on microtiter plates and used to
capture hu5A8-specific plasma antibodies. Wells were then treated with
HRP-conjugated hu5A8 capable of binding to the plasma immunoglobulin second
binding site. The specificity of the captured plasma immunoglobulin was assessed
by determining whether

<PAGE>
mu5A8 or control human IgG4 could inhibit binding of the HRP-conjugated hu5A8 to
captured plasma immunoglobulin. Inhibition was measured by routine ELISA similar
to that described above.

                                     RESULTS

hu5A8 RETAINS AFFINITY FOR HUMAN AND RHESUS MONKEY CD4

   Previously we had shown that mu5A8 had similar affinity for human and rhesus
monkey CD4.5 To determine whether humanization altered antigen recognition, we
compared the binding of mu5A8 and hu5A8 to a human CD4+ cell line using an
indirect immunofluorescence assay. Both MAbs bound to human CD4 with the
concentration yielding 50% maximal binding (EC50) of 10 to 25 ng/ml (Fig. 3A).
The ability of hu5A8 and mu5A8 to compete for binding to human CD4+ cells was
also equivalent (data not shown). To assess relative binding to monkey CD4 we
used a similar assay on monkey PBLs. As shown in Fig. 3B, the use of either the
murine or humanized antibody resulted in similar binding curves with an EC50 of
2 to 20 ng/ml providing an estimate for binding affinity of 10-100 pM. These
data indicate that the binding activity of humanized 5A8 for both human and
monkey CD4 is retained.

HU5A8 IS AS EFFECTIVE AS MU5A8 IN INHIBITING HIV-1-INDUCED SYNCYTIUM FORMATION
AND BLOCKIN SIVMAC REPLICATION in vitro

   To determine whether humanization had affected the potent IN VITRO antiviral
properties of this antibody, we assessed the ability of both mu5A8 and hu5A8 to
inhibit HIV-1-induced syncytium formation in an infectable T cell line. As shown
in Fig. 4, mu5A8 and hu5A8 were equally effective in blocking virally induced
syncytium formation between HIV-1-infected H9 cells and uninfected C8166 cells.
To compare the efficiency of mu5A8 and hu5A8 to inhibit SIVmac, we utilized a
virus outgrowth assay in which CD8-depleted PBLs from a chronically
SIVmac-infected monkey were cultured in the presence of either antibody or an
irrelevant isotype-matched murine MAb (Fig. 5). Both mu5A8 and hu5A8 completely
inhibited virus replication in these cultures at antibody concentrations of 1.0
and 10 Mg/ml. Partial inhibition of replication was observed at the 0.1-Mg/ml
concentration for both antibodies. High levels of replication were present in
cultures that contained MOPC-21, a control murine MAb, at all three
concentrations. Comparable results were observed when antiviral activity was
assessed against primary HIV-1 isolates (data not shown).

hu5A8 HAS PROLONGED PLASMA HALF-LIFE

   Single intravenous injections of hu5A8 were given to normal rhesus monkeys
and the disappearance of MAb from blood was measured as described. The results
in animals receiving 1, 3, and 30 mg of MAb per kilogram are shown in Fig. 6A.
The beta phase of clearance, which represents excretion rather than
redistribution, was 5.2 days for the 3-mg/kg dose and 5.6 days for the 30-mg/kg
dose. This was significantly longer than that observed for the murine form of
the antibody, which had a beta phase T1/2 of approximately 28 hr (Fig. 6B and
Ref. 5).

hu5A8 BINDS TO CIRCULATING AND LYMPH NODE CD4 CELLS WITHOUT INDUCING CLEARANCE

<PAGE>
   The original murine form of this antibody exhibited the favorable property IN
VIVO of binding to all potential target CD4+ cells without inducing significant
clearance of these coated cells. Coating of all circulating CD4+ lymphocytes
with MAb was sustained until the eventual appearance of the anti-mouse
immunoglobulin response when coating was no longer detected. Because the
constant regions of this newly engineered antibody were quite distinct from the
parent murine antibody, we wished to ascertain that this IN VIVO biological
effect was retained in hu5A8. CD4 cells were quantified in the blood of rhesus
monkeys before and periodically following a single intravenous injection of
hu5A8 at 1, 3, and 30 mg/kg. As illustrated in Fig. 7, all three dosage levels
resulted in coating of CD4+ lymphocytes and no clearance of CD4 cells was
observed during the 2-week period postadministration. In fact, at all dosage
levels, there was a transient increase in the absolute number of circulating
lymphocytes immediately after hu5A8 administration owing predominantly to an
increase in the number of circulating CD4 cells. This transient increase in CD4
cells was identical to the effect observed following administration of mu5A8 to
rhesus monkeys(5).

   To determine whether intravenous administration of hu5A8 could reach CD4+
cells within lymphoid organs, we performed lymph node biopsies 3 days after
administration of 3 mg of hu5A8 per kilogram body weight. Immunophenotypic
analyses showed that all CD4(+) lymphocytes within peripheral lymph nodes had
human IgG4 antibody on their surface (data not shown).

ANTIBODY RESPONSES ARISE AGAINST hu5A8 AND ARE ENTIRELY ANTIIDIOTYPE

   All monkeys that received 3-mg/kg doses of hu5A8 developed strong humoral
anti-hu5A8 responses at approximately 2 weeks postinjection. Similar results
were observed in monkeys that received the 30-mg/kg dose; three of four animals
developed antibodies against hu5A8 in approximately the same time frame. One
monkey at this dosage level failed to develop anti-hu5A8 antibodies. Of two
monkeys that received the 1-mg/kg dose, one failed to develop a response and the
other exhibited a delayed response 5 weeks post hu5A8 mAb. Typical responses
elicited in animals administered 3 mg/kg hu5A8 are illustrated in Fig. 8A. The
appearance of this anti-hu5A8 response corresponded to an abrupt loss of CD4
binding activity and was associated with a loss of coating of CD4+ lymphocytes
with hu5A8.

   To determine the fine specificity of this response, plasma from monkeys
treated with 3 mg of hu5A8 per kilogram were assayed by ELISA for reactivity
against hu5A8, mu5A8, and human IgG4. Representative results are shown in Fig.
8B. All plasma that recognized hu5A8 also reacted with an equivalent titer
against mu5A8. However, recognition of human IgG4 was never observed with any
plasma. The fine specificity of plasma antibodies from monkeys that received 3
mg of hu5A8 per kilogram was further confirmed using immobilized hu5A8 to
capture plasma antibodies that were capable of binding soluble HRP-conjugated
hu5A8 with their second antibody combining site. In these assays, murine 5A8
completely abolished binding of soluble hu5A8 to the monkey plasma antibodies.
By contrast, there was no effect of soluble human IgG4 on the binding hu5A8
(data not shown). Thus the major antibody response against hu5A8 was directed
against the murine V region determinants of the humanized antibody.

hu5A8 COATS CD4 CELLS IN SIV(mac)-INFECTED MONKEYS FOR >5 WEEKS

<PAGE>
   To determine the feasibility of using hu5A8 as a therapy in AIDS virus
infection, we administered weekly doses of hu5A8 or human IgG4 antibody at 3
mg/kg to rhesus monkeys chronically infected with SIV mac. We reasoned that
virally induced immunosuppression in these animals may permit a longer treatment
window before anti-hu5A8 antibodies developed. Plasma from all monkeys was
assayed for anti-hu5A8 activity prior to the weekly administrations. Antibodies
against hu5A8 were not detected in either of two monkeys (Fig. 9B) and treatment
was continued for a total of 6 weeks. Trough hu5A8 levels were maintained at 0.6
to 2.0 mg/ml for the entire 6-week period. Whereas circulating CD4+ lymphocyte
counts were lower in chronically SIVmac-infected monkeys than in normal monkeys,
administration of hu5A8 was not associated with any further CD4 cell depletion.
In fact, one of two treated monkeys showed a transient threefold increase in
circulating CD4+ cells (Fig. 9A). Studies were performed to ensure that CD4+
lymphocytes bound hu5A8 throughout the treatment period. Peripheral blood
lymphocytes from treated monkeys were dual labeled with FITC-conjugated
anti-human IgG4 and phycoerythrin-conjugated, noncompeting anti-CD4. Both
monkeys treated with hu5A8 maintained detectable humanized antibody on CD4+ PBLs
through the last treatment. Figure 10 illustrates this dual staining of
hu5A8-treated animals, but not control-treated or untreated animals on
experimental day 33. Other untoward effects of this extended treatment were not
observed.

                                   DISCUSSION

   Previously we explored the IN VIVO biological activity in rhesus monkeys of a
unique, CD4 domain 2-specific monoclonal antibody that was a potent inhibitor of
HIV-1 replication.5 In contrast to the action of many other T cell-specific MAbs
when administered to humans19-22 or nonhuman primates,23-26 this MAb was
nondepleting and failed to induce measurable immunosuppression. All CD4+
lymphocytes remained coated with this MAb until development of a monkey
anti-mouse immunoglobulin response. Here we report the successful humanization
of this MAb and explore its biological activity and potential utility IN VIVO
using nonhuman primates.

   The humanized form of this antibody retained its binding to human and rhesus
monkey CD4, as well as its potent IN VITRO antiviral activity against HIV-1 and
SIVmac. Like the parental murine MAb, the humanized form was nonmodulating and
nondepleting of circulating CD4+ lymphocytes. In addition, the plasma half-life
was significantly longer for the humanized MAb than for the murine MAb,
increasing its potential as a therapeutic agent.

   Despite a marked decrease in antigenicity following humanization, a humoral
response to hu5A8 did eventually develop in most normal monkeys and resulted in
loss of CD4 binding activity. However, the response was directed entirely
against V region determinants because no reactivity with human IgG4 was
observed. Responses showing a similar specificity for murine determinants have
been seen in macaque monkeys that received humanized forms of anti-CD4
(OKT4A),27 anti-CD25 (anti-Tac),28 or anti-human tumor necrosis factor A29
antibodies. Antibody responses to murine immunoglobulin, whether elicited in
humans or monkeys, have been difficult to block using conventional modes of
immunosuppression.30-32 However, preliminary results in other laboratories
indicate that some humanized MAb may be nonantigenic when administered to

<PAGE>
humans at high doses or in conjunction with immunosuppressive agents.(29,32-34)
Thus, chronic administration of a therapeutic, humanized MAb remains a realistic
goal.

   Interestingly, monkeys infected with the nonhuman primate AIDS virus, SIVmac,
failed to develop anti-hu5A8 antibodies. The lack of an idiotype-specific immune
response in these animals was probably due to the immunosuppressed state of
SIV-infected animals, where an AIDS-like disease is induced in macaque
monkeys.35 This allowed us an opportunity to evaluate the effect of chronic
hu5A8 administration in a relevant animal model of AIDS. Two SIVmac-infected
monkeys that received six weekly injections of hu5A8 maintained circulating
hu5A8 levels and hu5A8 coated CD4+ lymphocytes without any decrease in CD4+ cell
number throughout this entire period. Previously we showed that administration
of the murine form of 5A8 to SIVmac-infected monkeys could decrease provirus
load.36 On the basis of the data reported here, the humanized form of this
antibody could have distinct advantages over the murine MAb owing to its
decreased antigenicity and extended plasma half-life, and most importantly,
long-term administration could have a more profound antiviral effect.

   The potential immunomodulatory effects of chronically administered hu5A8 were
not directly assessed in these studied. However, the ability of the murine form
of 5A8 to block immunologic responses has been measured both IN VITRO and IN
VIVO. Although we have previously found that mu5A8 can exhibit some suppression
of antigen-specific proliferative responses IN VITRO, this inhibition was not
consistently observed (our unpublished results). By contrast, antibodies such as
OKT4A33 consistently showed strong inhibition in the same assays. Furthermore,
when mu5A8 was administered to rhesus monkeys, a dosage sufficient to coat all
CD4+ lymphocytes for 9 days failed to inhibit various T cell-dependent
responses.(5) In experiments performed to date, we find that the IN VITRO
activity of hu5A8 is similar to that of mu5A8, showing some inhibition of cell
proliferation in about half of antigen-stimulated cultures (our unpublished
results). However, any potential immunomodulatory effects of hu5A8 will be best
addressed by further IN VIVO analyses.

   Passive immunotherapy directed against CD4 has been proposed for both the
prevention and treatment of HIV-1 infection(37-40). This therapeutic modality
may have use as a monotherapy or may demonstrate more potent activity in
combination with conventional antiretroviral agents. IN VITRO experiments have
indicated that the antiviral activity of mu5A8 is synergistic with anti-HIV-1
gp120 antibodies.(41) The results reported here support the feasibility of using
CD4-directed passive immunotherapy for the treatment of AIDS.

                                 ACKNOWLEDGMENTS

   The authors thank Jim Estrella for performing the pharmacokinetic assays and
Brenda Emergian for executing the ELISA to detect monkey anti-hu5A8 antibodies.

   This work was supported by PHS Grants RR00168 and RR00055 and by funds from
Biogen, Inc.

<PAGE>
                                   REFERENCES

1.  Sattentau QJ and Weiss RA: The CD4 antigen: Physiological ligand and HIV
    receptor. Cell 1988;52:631-633.

2.  Dalgleish AG, Beverley PCL, Clapham PR, Crawford DH, Greaves MF, and Weiss
    RA: The CD4(T4) antigen is an essential component of the receptor for the
    AIDS retrovirus. Nature (London) 1984;312:763-767.

3.  Klatzman D, Champagne E, Chamaret S, Gruest J, Guetard D, Hercend T,
    Gluckman J-C, and Montagnier L: T-lymphocytes T4 molecule behaves as the
    receptor for human retrovirus LAV. Nature (London) 1984;312:767-768.

4.  McDougal JS, Kennedy MS, Sligh JM, Cort SP, Mawle A, and Nicholson JKA:
    Binding of HTLV-III/LAV to T4+ T cells by complex of the 110K viral protein
    and the T4 molecule. Science 1986;231:382-385.

5.  Reimann KA, Burkly LC, Burrus B, Waite BCD, Lord CI, and Letvin NL: IN VIVO
    administration to rhesus monkeys of a CD4-specific monoclonal antibody
    capable of blocking AIDS virus replication. AID Res Hum Retroviruses
    1993;9:199-207.

6.  Burkly LC, Olson D, Shapiro R, Winkler G, Rosa JJ, Thomas DW, Williams C,
    and Chisholm P: Inhibition of HIV infection by a novel CD4 domain 2-specific
    monoclonal antibody. Dissecting the basis for its inhibitory effect on
    HIV-induced cell fusion. J Immunol 1992;149:1779-1787.

7.  Moore JP, Sattentau QJ, Klasse PJ, and Burkly LC: A monoclonal antibody to
    CD4 domain 2 blocks soluble CD4-induced conformational changes in the
    envelope glycoproteins of human immunodeficiency virus type 1 (HIV-1) and
    HIV-1 infection of CD4+ cells. J Virol 1992;66:4784-4793.

8.  Orlandi R, Gussow DH, Jones PT, and Winter G: Cloning immunoglobulin
    variable domains for expression by the polymerase chain reaction. Proc Natl
    Acad Sci USA 1989;86:3833-3837.

9.  Kabat EA, Wu TT, Reid-Miller M, Perry HM, and Gottesman KS: SEQUENCES OF
    PROTEINS OF IMMUNOLOGICAL INTEREST, Vol. 4, 5th Ed. U.S. Department of
    Health and Human Services, Bethesda, Maryland, 1991.

10. Klobeck HG, Bornkamm GW, Combriato G, Mocikat R, Pohlenz HD, and Zachau HG:
    Subgroup IV of human immunoglobulin k light chains is encoded by a single
    germline gene. Nucleic Acids Res 1985;13:6515-6529.

11. Larrick JW, Danielsson L, Brenner CA, Abrahamson M, Fry KE, and Borrebaeck
    CA: Rapid cloning of rearranged immunoglobulin genes from human hybridoma
    cells using mixed primers and polymerase chain reaction. Biochem Biophys Res
    Commun 1989;160:1250-1256.

12. Padlan EA: Anatomy of the antibody molecule. Mol Immunol 1994;31:169-217.

13. Bebbington CR, Renner G, Thomson S, King D, Abrams D, and Yarranton GT: High
    level expression of a recombinant antibody from myeloma cells using a
    glutamine synthetase gene as an amplifiable selectable marker. Biotechnology
    1992;10:169-175.

14. Galfre G and Milstein C: Preparation of monoclonal antibodies: Strategies
    and procedures. Methods Enzymol 1981;73:3-46.

15. National Institutes of Health: GUIDE FOR THE CARE AND USE OF LABORATORY
    ANIMALS, Rev. Ed. Department of Health and Human Services Publication No.
    (NIH) 85-23. National Institutes of Health, Bethesda, Maryland, 1985.

<PAGE>
16. Walker BD, Kowalski M, Goh WC, Kozarsky K, Krieger M, Rosen C, Rohrschneider
    L, Haseltine WA, and Sodroski J: Inhibition of human immunodeficiency virus
    syncytium formation and virus replication by castanospermine. Proc Natl Acad
    Sci USA 1987;84:8120-8124.

17. Reimann KA, Waite BCD, Lee-Parritz DE, Lin W, Uchanska-Ziegler B, O'Connell
    MJ, and Letvin NL: Use of human leukocyte-specific monoclonal antibodies for
    clinically immunophenotyping lymphocytes of rhesus monkeys. Cytometry
    1994;17:102-108.

18. Cobbold SP, Rebello PRUB, Davies HFS, Friend PJ, and Clark MR: A simple
    method for measuring patient anti-globulin responses against isotopic or
    idiotypic determinants. J Immunol Methods 1990;127:19-24.

19. Norman DJ: Mechanisms of action and overview of OKT3. Ther Drug Monitoring
    1995;17:615-620.

20. Shizuru JA, Alters SE, and Fathman CG: Anti-CD4 monoclonal antibodies in
    therapy: Creation of nonclassical tolerance in the adult. Immunol Rev
    1992;129:105-130.

21. Riethmuller G, Rieber EP, Kiefersauer S, Prinz J, van der Lubbe P, Meiser B,
    Breedveld F, Eisenberg J, Kruger K, Deusch K, ET AL.: From antilymphocyte
    serum to therapeutic monoclonal antibodies: First experiences with a
    chimeric CD4 antibody in the treatment of autoimmune disease. Immunol Rev
    1992;129:81-104.

22. Moreland LW, Bucy RP, Tilden A, Pratt PW, LoBuglio AF, Khazaeli M, Everson
    MP, Daddona P, Ghrayeb J, Kilgarriff C, Sanders ME, and Koopman WJ: Use of a
    chimeric monoclonal anti-CD4 antibody in patients with refractory rheumatoid
    arthritis. Arthritis Rheumatism 1993;36:307-318.

23. Rose LM, Alvord EC, Hruby S, Jackevicius S, Petersen R, Warner N, and Clark
    EA: IN VIVO administration of anti-CD4 monoclonal antibody prolongs survival
    in longtailed macaques with experimental allergic encephalomyelitis. Clin
    Immunol Immunopathol 1987;45:405-423.

24. Jonker M, Goldstein G, and Balner H: Effects of IN VIVO administration of
    monoclonal antibodies specific for human T cell subpopulations of the immune
    system in a rhesus monkey model. Transplantation 1983;35:521-526.

25. Cosimi AB, Burton RC, Kung PC, Colvin R, Goldstein G, Lifter J, Rhodes W,
    and Russell PS: Evaluation in primate renal allograft recipients of
    monoclonal antibody to human T-cell subclasses. Transplantation Proc
    1981;13:499-503.

26. Jonker M, Slingerland W, Treacy G, van Eerd P, Pak KY, Wilson E, Tam S,
    Bakker K, LoBuglio AF, Rieber P, ET AL.: IN VIVO treatment with a monoclonal
    chimeric anti-CD4 antibody results in prolonged depletion of circulating
    CD4+ cells in chimpanzees. Clin Exp Immunol 1993;93:301-307.

27. Delmonico FL, Cosimi AB, Kawai T, Cavender D, Lee W-H, Jolliffe LK, and
    Knowles RW: Nonhuman primate responses to murine and humanized OKT4A.
    Transplantation 1993;55:722-728.

28. Schneider WP, Glaser SM, Kondas JA, and Hakimi J: The anti-idiotypic
    response by cynomolgus monkeys to humanized anti-Tac is primarily directed
    to complementarity-determining regions H1, H2, and L3. J Immunol
    1993;150:3086-3090.

29. Stephens S, Emtage S, Vetterlein O, Chaplin L, Bebbington C, Nesbitt A,
    Sopwith M, Athwal D, Novak C, and Bodmer M: Comprehensive pharmacokinetics
    of a humanized antibody and analysis of residual anti-idiotypic responses.
    Immunology 1995;85:668-674.

<PAGE>
30. Jaffers GJ, Fuller TC, Cosimi AB, Russell PS, Winn HJ, and Colvin RB:
    Monoclonal antibody therapy: Anti-idiotypic and non-anti-idiotyic antibodies
    to OKT3 arising despite intense immunosuppression. Transplantation
    1986;41:572-578.

31. Lowder JN, Miller RA, Hoppe R, and Levy R: Suppression of anti-mouse
    immunoglobulin antibodies in subhuman primates receiving murine monoclonal
    antibodies against T cell antigens. J Immunol 1987;138:401-406.

32. Khazaeli MB, Conry RM, and LoBuglio AF: Human immune response to monoclonal
    antibodies. J Immunother 1994;15:42-52.

33. Pulito VL, Roberts VA, Adair JR, Rothermel AL, Collins AM, Varga SS,
    Martocello C, Bodmer M, Jolliffe LK, and Zivin RA: Humanization and
    molecular modeling of the anti-CD4 monoclonal antibody. OKT4A. J Immunol
    1996;156:2840-2850.

34. Anasetti C, Hansen JA, Waldmann TA, Appelbaum FT, Davis J, Deeg HJ, Doneyu
    K, Martin PJ, Nash R, Storb R, and Strom TA: Treatment of acute
    graft-versus-host disease with humanized anti-Tac: An antibody that binds to
    the interleukin-2 receptor. Blood 1994;84:1320-1327.

35. Letvin NL and King NW: Immunologic and pathologic manifestations of the
    infection of rhesus monkeys with simian immunodeficiency virus of macaques.
    J Acquir Immune Defic Syndr 1990;3:1023-1040.

36. Reimann KA, Cate RL, Wu Y, Palmer L, Olson D, Waite BCD, Letvin NL, and
    Burkly LC: IN VIVO administration of CD4-specific monoclonal antibody:
    Effect on provirus load in rhesus monkeys chronically infected with the
    simian immunodeficiency virus of macaques. AIDS Res Hum Retroviruses
    1995;11:517-525.

37. Attanasio R, Allan JS, and Kennedy RC: Monoclonal anti-CD4 as
    immunoprophylactic agents for human immunodeficiency virus infection. J
    Infect Dis 1993;168:515-516.

38. Rieber EP, Federle C, Reiter C, Krauss S, Gurtler L, Eberle J, Deinhardt F,
    and Riethmuller G: The monoclonal CD4 antibody M-T413 inhibits cellular
    infection with human immunodeficiency virus after viral attachment to the
    cell membrane: An approach to postexposure prophylaxis. Proc Natl Acad Sci
    USA 1992;89:10792-10796.

39. Dhiver C, Olive D, Rousseau S, Tamalet C, Lopez M, Galindo J-R, Mourens M,
    Hirn M, Gastaut J-A, and Mawas C: Pilot phase I study using zidovudine in
    association with a 10-day course of anti-CD4 monoclonal antibody in seven
    AIDS patients. AIDS 1989;3:835-842.

40. Reiber EP, Reiter C, Guntler L, Deinhardt F, and Riethmuller G: Monoclonal
    CD4 antibodies after accidental HIV infection. Lancet 1990;336:1007-1008.

41. Burkly L, Mulrey N, Blumenthal R, and Dimitrov DS: Synergistic inhibition of
    human immunodeficiency virus type 1 envelope glycoprotein-mediated cell
    fusion and infection by an antibody to CD4 domain 2 in combination with
    anti-gp120 antibodies. J Virol 1995;69:4267-4273.

                                                    Address reprint requests to:
                                                                KEITH A. REIMANN
                                                  DIVISION OF VIRAL PATHOGENESIS
                                            BETH ISRAEL DEACONESS MEDICAL CENTER
                                                                          RE-113
                                                            330 BROOKLINE AVENUE
                                                     BOSTON, MASSACHUSETTS 02215

<PAGE>
(1) Division of Viral Pathogenesis, Department of Medicine, Beth Israel
Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts 02215.

(2) Biogen, Inc., Cambridge, Massachusetts 02142.

(3) Laboratory of Molecular Biology, National Institute of Diabetes and
Digestive and Kidney Diseases, National Institutes of Health, Bethesda, Maryland
20892.

FIG. 1. Amino acid sequences of the V(H) and V(L) regions of mu5A8 and hu5A8,
and human antibody V(H) (MO30) and V(L) (VJI). MO30 and VJI contain the human
framework sequences most homologous to the murine V(H) and V(L) sequences,
respectively. The sequences are displayed in segments revealing those regions
corresponding to the four framework (FR) and three complementarity-determining
regions (CDRs). Dots in the hu5A8 VH and VL sequences indicate identity with
MO30 and VJI, respectively.

FIG. 2.  Oligonucleotides used in this study.

FIG. 3. hu5A8 and mu5A8 bind similarly to human and rhesus monkey CD4. The
relative binding abilities of hu5A8 and mu5A8 to human and rhesus monkey CD4+
cells were compared. A human CD4+ cell line (A) and rhesus monkey PBLs (B) were
incubated with increasing concentrations of either antibody and binding to cells
determined by indirect immunofluorescence. (F) mu5A8; ( ) hu5A8.

FIG. 4. hu5A8 inhibits syncytium formation in HIV-1-infected cells as well as
mu5A8. H9 cells chronically infected with HIV-1 were cocultured with uninfected
C8166 cells in the presence of varying concentrations of mu5A8 (F), hu5A8 (..),
or a control murine immunoglobulin ( ). Syncytia were counted in each well and
expressed as percentage inhibition by comparing values with control
immunoglobulin-treated wells.

FIG. 5. hu5A8 suppresses SIVmac replication as efficiently as the parental
murine monoclonal antibody. PBLs were isolated from the blood of a monkey
infected with SIVmac, activated with concanavalin A, and cultured in the
presence of hu5A8, mu5A8, or an isotype-matched control murine antibody
(MOPC-21) at 0.1 mg/ml (F), 1 mg/ml (F), or 10 mg/ml (G). SIVmac p27Gag antigen
was quantitated in the culture supernatant at four time points after initiating
cultures.

FIG. 6. Plasma half-life of hu5A8 is significantly longer than mu5A8. (A) A
single intravenous injection of hu5A8 was administered on day 0 to normal rhesus
monkeys at 1 mg/kg (F), 3 mg/kg (..), or 30 mg/kg (F). Plasma levels were
determined serially thereafter. (B) Clearance of hu5A8 (..) and mu5A8 ( ) at 3
mg/kg are compared. Each line represents an individual monkey.

FIG. 7. Administration of hu5A8 results in a transient increase in circulating
CD4+ lymphocytes. Circulating CD4+ lymphocytes were enumerated in the blood of
normal rhesus monkeys using OKT4-PE after a single intravenous injection of
hu5A8 at the indicated dose. Each line represents an individual animal. Arrow
indicates time of injection on day 0. Bar indicates time when cells coated with
hu5A8 were detected using anti-human IgG4-FITC.

<PAGE>
FIG. 8. Monkeys develop an anti-hu5A8 response that is predominantly
antiidiotype. (A) Two normal rhesus monkeys that received 3 mg of hu5A8 per
kilogram body weight developed antibodies that recognized hu5A8 approximately 2
weeks after a single intravenous injection. Monkey 1 (G); monkey 2 ( ). (B) Fine
specificity of antibodies in the same two hu5A8-treated monkeys. Plasma
recognized hu5A8 and parental mu5A8, but not another human IgG4. All values
correspond to the highest dilution that gave a positive signal as measured by
ELISA.

FIG. 9. Monkeys infected with simian immunodeficiency virus of macaques
(SIV(mac)) could be treated with hu5A8 for 6 weeks without developing anti-hu5A8
antibodies. Two SIV(mac)-infected rhesus monkeys received weekly intravenous
injections (indicated by arrows) of 3 mg of hu5A8 per kilogram body weight. (A)
CD4+ lymphocyte counts in peripheral blood of two treated monkeys (.,o,o) showed
no loss of target cells. Values from an untreated, SIVmac-infected animal are
also shown ( ). (B) Anti-hu5A8 antibody titers remained low in plasma of treated
monkeys assayed weekly during hu5A8 treatment. As a positive control, plasma
from a similarly treated, uninfected animal was assayed in parallel.

FIG. 10. Humanized 5A8 continues to binds to circulating CD4+ lymphocytes of
SIV(mac)-infected rhesus monkeys for more than 4 weeks during continuous
treatment. SIV(mac)-infected rhesus monkeys received 3 mg of hu5A8 or a control
antibody per kilogram body weight weekly for 3 weeks as described in Fig. 9.
Circulating lymphocytes were assessed for surface-bound hu5A8 on experimental
day 33 by indirect immunofluorescence with FITC-conjugated anti-human IgG4. Cell
were also colabeled with a phycoerythrin-conjugated, noncompeting anti-CD4
antibody (OKT4-PE) and fluorescence determined flow cytometrically. (A)
Untreated control animal; (B) human IgG4 control treated; (C and D) two
hu5A8-treated animals.

<PAGE>
AIDS RESEARCH AND HUMAN RETROVIRUSES
Volume 9, Number 3, 1993
Mary Ann Liebert, Inc., Publishers

      IN VIVO ADMINISTRATION TO RHESUS MONKEYS OF A CD4-SPECIFIC MONOCLONAL
               ANTIBODY CAPABLE OF BLOCKING AIDS VIRUS REPLICATION
       KEITH A. REIMANN,* LINDA C. BURKLY,+ BEATRICE BURRUS,+ BARRY C. D.
                  WATTE,* CAROL I. LORD,* and NORMAN L. LETVIN*

ABSTRACT

Monoclonal antibodies (mAbs) specific for CD4 are potent inhibitors of HIV
replication IN VITRO. These agents may be useful prophylactically or in chronic
HIV infection if they can be administered without inducing immunosuppression. In
the present study, we explored the safety of a CD4-specific murine mAb in rhesus
monkeys. The mAb 5A8, which binds to domain 2 of the CD4 molecule, inhibits AIDS
virus replication noncompetitively at a postvirus binding step. This antibody,
which had a similar affinity for rhesus monkey and human CD4 cells, efficiently
inhibited IN VITRO replication of both HIV-1 and the simian immunodeficiency
virus of macaques. A single 3-mg/kg injection of mAb 5A8 into normal rhesus
monkeys coated all circulating and lymph node CD4 cells for 4-6 days. CD4 cells
were not cleared from circulation nor was the CD4 molecule modulated from the
lymphocyte surface. In fact, administration of mAb 5A8 resulted in an
approximately one- to twofold increase in absolute number of circulating CD4
cells. Repeated administration in normal rhesus monkeys resulted in CD4
lymphocyte coating with mAbs for greater than 9 days without CD4 cell clearance
or modulation. While coated with mAbs, PBLs of these monkeys retained normal IN
VITRO proliferative responses to mitogens and these animals generated normal
humoral responses IN VIVO to tetanus toxoid. Loss of cell coating with mAbs in
normal monkeys corresponded to the appearance of anti-mouse immunoglobulin
antibodies. Thus, administration of certain anti-CD4 mAbs capable of blocking
HIV replication can achieve coating of the entire CD4 cell pool in rhesus
monkeys without inducing significant cell loss or immunosuppression.

                                  INTRODUCTION

THE CD4 MOLECULE IS A high-affinity receptor for the envelope glycoprotein of
HIV.(1-3) Thus, certain anti-CD4 monoclonal antibodies (mAbs) can block
HIV-induced syncytium formation and infection of lymphocytes and macrophages IN
VITRO.(1-3) The antiviral activity of various anti-CD4 mAbs has been ascribed to
steric interference with virus-CD4 binding,(1,2) inhibition of postbinding
conformational changes in CD4 required for fusion,(4,5) and CD4-mediated
inhibition of cell activation.(6)

      Monoclonal antibodies are being assessed as therapeutic agents in a
variety of clinical settings: as cancer therapies,(7,8) for immunomodulation in
autoimmune disorders,(9) in blocking allograft rejection,(9,10) and to bind
endotoxin in gram-negative bacteremia.(11,12) Yet in spite of their ability to
inhibit HIV replication IN VITRO, mAbs that bind CD4 have received little
attention as a potential AIDS therapy.(13) We(14,15) and others(16) have
demonstrated that antibodies to self-CD4 can be
<PAGE>
generated IN VIVO without inducing immune dysfunction. We therefore wished to
explore the feasibility of using anti-CD4 mAbs as passive immunotherapy that may
have utility in HIV prophylaxis or therapy.

      We have employed rhesus monkeys in this study because many human
leukocyte-specific monoclonal antibodies cross-react with conserved
determinants, including CD4, on monkey cells.(17,18) In addition, the simian
immunodeficiency virus of macaques (SIVmac), a lentivirus similar to HIV both
genetically and in its CD4 cell tropism, causes an AIDS-like disease in rhesus
monkeys.(19,20) The phylogenetic proximity of humans to the rhesus monkey and
the similarities of the immunopathogenesis of SIV(mac)-induced disease in the
monkey and HIV-induced disease in humans make this animal model uniquely suited
to evaluate immune-based therapies for AIDS. We selected the anti-CD4 mAb 5A8
for study IN VIVO because it is a potent inhibitor of HIV infection and cell
fusion.(21) Moreover, 5A8 may prove more useful as a therapeutic agent than
other anti-CD4 mAbs because of its ability to inhibit HIV infection/fusion in a
noncompetitive fashion, blocking postvirus/CD4 binding steps of
infection.(21,22) We sought to determine whether this anti- CD4 mAb could be
administered parenterally to rhesus monkeys without compromising immune
function.

                              MATERIALS AND METHODS

ANIMALS

      The rhesus monkeys (MACACA MULANA) used in this study were maintained in
accordance with the guidelines of the Committee on Animals for the Harvard
Medical School and the GUIDE FOR THE CARE AND USE OF LABORATORY ANIMALS
(Department of Health and Human Services Publication No. NIH 85-23, revised
1985). Monkeys were anesthetized with ketamine-HCl for all procedures.

PREPARATION OF MONOCLONAL ANTIBODIES

      The anti-CD4 mAb 5A8 was isolated and characterized as described
previously.(21,22) 5A8 and an isotype-matched control mAb. MOPC-21, were
purified by protein A affinity chromatography from tissue culture supernatants.
Purified mAbs contained only 2.2 and 1.4 endotoxin units/mg for 5A8 and MOPC-21,
respectively.

CULTURING OF HUMAN AND MONKEY CELLS

      Blood samples from normal rhesus monkeys and from normal human donors were
collected in heparinized syringes and peripheral blood lymphocytes (PBLs) were
isolated by density gradient centrifugation. Human concanavalin A (Con A) blasts
were prepared by incubating normal PBLs in medium with 5 mg of Con A (Sigma
Chemical Co., St. Louis, MO) per milliliter for 12-24 hr. H9 cells were obtained
from NIH AIDS Research and Reference Reagent Program (Rockville, MD). Cells were
cultured routinely in complete medium, which consisted of RPMI 1640 supplemented
with penicillin (50 U/ml), gentamicin (50 mg/ml), N-2-
hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES) buffer (10 mM),
L-glutamine (2 mM), and 10% heat-inactivated fetal bovine serum

<PAGE>
(ICN, Inc., Costa Mesa, CA) at 37(degree)C in a 5% Co(2), humidified atmosphere.
All human PBL cultures were supplemented with 20 U/ml of human recombinant
interleukin 2 (IL-2) (Hoffman-La Roche, Nutley, NJ).

AFFINITY STUDIES

      Isolated PBLs were incubated with varying concentrations of mAb 5A8
diluted in phosphate-buffered saline (PBS)-0.1% bovine serum albumin (BSA) (2 x
10(4) cells/100 ml) for 20 min at room temperature. Cells were then washed and
incubated with goat anti-mouse IgG- fluorescein isothiocyanate (FITC) [F(ab)(2)
fragment] for 10 min at room temperature, washed again, and fixed with PBS-1%
formalin. Mean fluorescence of the CD4+ PBLs was determined for each sample by
flow cytometry and binding curves for monkey and human cells were compared.

IN VITRO BLOCKING OF HIV AND SIV REPLICATION

      The ability of the mAb 5A8 to block AIDS virus infection of cells IN VITRO
was assessed by incubating either H9 cells with SIV(mac), or human PBL Con A
blasts with HIV-1(MN), in the continuous presence of 5A8 or an isotype-matched
control antibody. After a 12-hr incubation with virus, cells were washed once
and cultured as described above. Every 3 to 4 days an aliquot of medium was
collected to assay for viral antigen and the cell number in the cultures was
adjusted to approximately 5 x 10(5)/ml with fresh medium.

      To determine whether 5A8 could block virus outgrowth from PBLs of
SIV(mac)-infected monkeys or HIV-infected humans, mononuclear cells were
isolated from the blood of infected subjects and activated overnight with 5 mg
of Con A per milliliter of complete medium. CD8 cells were then depleted from
these cultures by using an anti-CD8 antibody (7PT3F9, provided by Dr. S.
Schlossman, Dana Farber Cancer Institute, Boston, MA) and immunomagnetic beads.
These CD8- depleted cultures were then cultured routinely in the presence of 5A8
as described above. Culture medium collected every 3 to 4 days was assayed for
the presence of either SIV p27 or HIV p24 antigen, using commercially available
kits (SIV core antigen assay, HIV-1 core antigen assay, Coulter Corp., Hialeah,
FL).

INTRAVENOUS ADMINISTRATION OF MONOCLONAL ANTIBODIES

      Monkeys were anesthetized and administered 3 mg of mAb 5A8 per kilogram or
a control antibody diluted to 1-2 mg/ml in PBS by intravenous injection over
approximately 1 min. In experiments in which multiple administrations were used,
this same dosage was given every 2 days.

IMMUNOPHENOTYPING BLOOD AND LYMPH NODE LYMPHOCYTES

      Biopsied lymph nodes were placed in complete medium and gently teased with
forceps to place cells in suspension. This suspension was then filtered through
112-mm nylon mesh and the cell number was adjusted to 2 x 10(6) nucleated cells
per milliliter in Hanks' balanced salt solution with 2.5% fetal bovine serum.
Both the lymph node lymphocytes and PBLs were stained with cell- specific mAbs
and analyzed by flow cytometry.
<PAGE>
      Ethylenediaminetetraacetic acid (EDTA)-anticoagulated blood was washed
three times with PBS-0.1% BSA to remove plasma and then cells were resuspended
in PBS-0.1% BSA at the original blood volume. Fifty microliters of blood cell or
lymph node cell suspension was added to 100 ml of PBS-0.1% BSA that contained an
appropriate amount of mAb. Monoclonal antibodies used were anti-CD4 [OKT4-FITC
(Ortho Diagnostic Systems, Raritan, NJ)], anti-CD8 [T8-FITC (Dako Corporation,
Carpinteria, CA)], and anti-CD45 [LCA (Dako Corp.)]. Cells were incubated for 30
min at room temperature and then washed once with PBS. To determine the extent
of IN VIVO cell coating with mAb 5A8 and in the stains for CD45, an indirect
staining technique was used. Cells were incubated, additionally, with a 1:50
dilution of FTTC-conjugated goat anti-mouse immunoglobulin [F(ab)2 fragment]
(Jackson ImmunoResearch, West Grove, PA) in PBS-0.1% BSA for 20 min at room
temperature and then washed with PBS. Red blood cells in all samples were lysed
by using a commercial RBC lysing kit (Immunolyse; Coulter Corp., Hialeah, FL),
washed in PBS, and resuspended in 0.4 ml of PBS-1% formalin. Samples were
analyzed routinely on an Epics-C flow cytometer, (Coulter Corp., Hialeah, FL)
using forward light scatter, 90(degree) light scatter, and CD45 fluorescence to
identify lymphocytes.

TETANUS TOXOID ANTIBODY TITERS

      Ninety-six-well microtiter plates (Immunolon: Dynatech, Chantilly, CA)
were coated with tetanus toxoid (Massachusetts State Laboratory, Boston, MA) at
5 m/ml in PBS, 50 ml/well overnight at 4(degree)C. Plates were washed with
PBS-0.5% Tween 20 and wells were blocked with PBS- 2% nonfat dry milk, 200
ml/well for 2 hr at room temperature. Wells were washed again and monkey serum
samples diluted in PBS-2% milk were added at 50 ml/well and incubated for 2-3 hr
at room temperature. After washing, horseradish peroxidase (HRP)-conjugated
anti-human immunoglobulin (Zymed Laboratories, San Francisco, CA) or
HRP-conjugated goat anti-human IgG Fc (Jackson ImmunoResearch, West Grove, PA)
was added at 50 ml/well for 1-2 hr at room temperature. After final washes,
O-phenyl-diamine substrate (Calbiochem, La Jolla, CA) was added at 100 ml/well
for 20 min. the reaction stopped with 1 N H2SO4 (100 ml/well), and absorbance
read at 490 nm on a microplate reader. Data are expressed as units total
immunoglobulin per milliliter, or IgG per milliliter, relative to a polyclonal
immune serum standard.

ANTI-MOUSE IMMUNOGLOBULIN LEVELS

      Microtiter plates were coated with mAb 5A8 at 5 mg/ml in bicarbonate
buffer (pH 9.0) overnight at 4(degree)C, washed with PBS, blocked with PBS-2%
milk, washed again, and monkey serum added as described above. Anti-mouse
immunoglobulin anti-bodies were detected with HRP- conjugated goat anti-human
immunoglobulin. Data are expressed as units immunoglobulin per milliliter
relative to a polyclonal immune serum standard.

MOUSE IMMUNOGLOBULIN SERUM LEVELS

      Microtiter plates were coated with goat anti-mouse IgG
(Cappel/Organon-Teknika Corp., Westchester, PA) at 5 mg/ml in bicarbonate buffer
(pH 9.0), blocked with PBS-2% milk, and monkey serum added at 50 ml/well as
described above. Mouse immunoglobulin was detected with

<PAGE>
HRP-conjugated rabbit and anti-mouse IgG (Cappe1/Organon-Teknika). Data are
expressed as micrograms mouse immunoglobulin per milliliter, using purified mAb
5A8 as a standard.

IN VITRO PROLIFERATION ASSAYS

      The proliferative responses of PBLs to pokeweed mitogen (PWM) and in a
mixed lymphocyte reaction (MLR) were measured at various time points in monkeys
that received mAb. Ficoll-isolated PBLs were washed and cultured in triplicate
in 96-well, round-bottomed polystyrene plates with 10(5) cells/well in 200
microliters of complete medium. For PWM proliferations, wells were supplemented
with PWM (GIBCO Bethesda Research Laboratories, Grand Island, NY) at a final
dilution of 1:300. For the MLR, 10(3) mitomycin C-treated B cells from a
xenogeneic Epstein-Barr virus (EBV)-transformed cell line was added to each
well. Peripheral blood lymphocytes cultured in medium alone served as
unstimulated controls. Cultures were maintained at 37 (degrees) C in a
humidified atmosphere with 5% CO(2). After 3 days for PWM-stimulated cultures
and 4 days for the MLR, cells were pulse labeled with 1 microliter of
[METHYL-(3)H]THYMIDINE for 4 hr. Cells were then harvested by an automated cell
harvester and thymidine incorporation was quantified by standard liquid
scintilography.

                                     RESULTS

      THE ANTI-CD4 mAb 5A8 HAD SIMILAR AFFINITY FOR MONKEY AND HUMAN CD4(+)
CELLS AND BLOCKED HIV AND SIV (MAC) REPLICATION IN VITRO. The HIVs and SIVs must
bind to membrane-expressed CD4 molecules to initiate a cellular infection. To
assess the ability ofmAb 5A8 to block these infections, the affinity of 5A8 for
monkey and human CD4 cells was first compared. The increase in fluorescence that
occurred as mAb concentration increased was similar for human and monkey CD(4+)
PBLs (Fig. 1 ). The concentrations of 5A8 that yielded 50% saturation of binding
sites were similar, approximately 0.8 micrograms/ml for human PBLs and 0.6
micrograms/ml for monkey PBLs.

      To assess directly the ability of this mAb to block AIDS virus infection.
CD4-bearing cells were inoculated with either SIV(mac) or HIV-(1)(MN) in the
presence of this antibody. As shown in Table 1, blocking of both HIV and SIV
infection was complete at an antibody concentration of 10 micrograms/ml. Partial
blocking was seen at 1 and 0.1 micrograms/ml. Outgrowth of virus from
CD8-depleted PBLs from HIV-l-infected humans and SIV (max)-infected monkeys was
completely inhibited at 10 and 1 micrograms/ml (Table 1). Thus, 5A8 can inhibit
IN VITRO infection and replication of both HIV and SIV with similar efficiency.

      PARENTERALLY ADMINISTERED mAb 5A8 RESULTED IN COATING OF CD4 CELLS WITHOUT
CLEARANCE FROM CIRCULATION. To determine the effect that parenteral
administration of this antibody would have on circulating CD4 lymphocytes, two
normal rhesus monkeys received a single intravenous administration of mAb 5A8.
As shown in Fig. 2A, the administration of 3 mg of antibody per kilogram
resulted in peak serum mouse immunoglobulin levels of 40-50 micrograms/ml.
Monoclonal antibody 5A8 remained detectable in the serum for at least 4 days in
both monkeys. The pharmacokinetics of its disappearance were consistent with a
two-phase clearance of mAb, with an (alpha) phase T (1/2) of approximately 2 hr
and a (Beta) phase T (1/2) of approximately 28 hr.
<PAGE>
      Following antibody administration, there was no evidence of a decrease in
percentage (Fig. 2C) or absolute number (Fig. 2B) of CD4(+) PBLs. In fact, 2
days following mAb treatment, total blood lymphocytes increased from 1600 to
2500/microliters in one and from 3100 to 6900/microliters in the other monkey.
This was due largely to an increase in circulating CD4 cells, which rose
modestly in one monkey (800 to 1400/microliters) and markedly in the other (1100
to 3000/micro1iters) (Table 2). The increase in circulating CD4 cell number
persisted as long as cell coating with mAb was evident By 5-6 days following
treatment, circulating lymphocytes had returned to pretreatment levels in both
monkeys, although one developed a CD4 lymphocytosis again in the second week
following treatment There were no significant changes in the other peripheral
Dlood leukocytes in treated monkeys.

      The single administration of anti-CD4 antibody resulted in coating of all
CD4 cells in the monkeys for a period of 4-6 days (Fig. 2D). As expected,
monkeys that received mAb developed humoral responses to the foreign mouse
protein (Fig. 2E). The appearance of this response appeared to correspond with
the disappearance of mouse immunoglobulin from the surface of monkey CD4
lymphocytes. No clinical signs were associated with the emergence of this
anti-mouse immunoglobulin response.

      ADMINISTERED mAb 5A8 COATED ALL CD4 CELLS IN SECONDARY LYMPHOID ORGANS.
Our previous studies with other T cell-specific mAbs in monkeys showed that
antibody dosages of at least 2 mg/kg were necessary to coat lymphocytes in
secondary lymphoid tissues.(23) To ensure that all CD4 lymphocytes in lymphoid
organs were coated with mAb 5A8, biopsies of peripheral lymph nodes were
performed in those two monkeys that received the single 3-mg/kg injection of
antibody. Lymph node lymphocytes from these biopsies were assessed by flow
cytometry for the presence of mouse immunoglobulin on their cell surface before
and after IN VITRO addition of mAb 5A8. No difference in the percentage of cells
staining positively following addition of mAb 5A8 indicated ; that all CD4 1ymph
node cells had been coated IN VIVO. As shown in Fig. 3, all CD4 cells were
coated with mouse mAb for 5 days after this single treatment.

      PERIPHERAL BLOOD LYMPHOCYTES OF MONKEYS THAT RECEIVED REPEATED
ADMINISTRATIONS OF 5AB RETAINED A NORMAL IN VITRO PROLIFERATIVE CAPACITY AND THE
MONKEYS GENERATED NORMAL ANTIBODY RESPONSES. We next examined the effects of
longer-term coating with mAb 5A8 on CD4 cell function. Two normal rhesus monkeys
were given intravenous mAb 5A8 (3 mg/kg) every other day for three treatments.
As controls, an additional two monkeys received an isotype-matched irrelevant
mAb, using the same dosing regimen. These administrations resulted in peak
plasma mouse immunoglobulin levels of 80-100 micrograms/ml in the 5A8-treated
and 250-300 microgram/ml in the control-treated monkeys, with detectable levels
remaining for at least 9 days.

      Circulating CD4 cells in the 5A8-treated monkeys remained coated with
mouse antibody for at least 9 days. The effect of repeated mAb 5A8
administration on circulating cell number was similar to the effect observed in
the monkeys receiving single injections. Increases in absolute number of
lymphocytes were observed, with a preferential increase reflected in the CD4(+)
subset. The increase in CD4(+) PBLs correlated with cell coating with mAb (Table
2, Fig. 4). Monkeys receiving the control mAb showed no consistent change in the
number of both total and CD4(+) PBLs over this period. Thus, prolonged coating
of circulating CD4 cells with mAb 5A8 did not result in their
<PAGE>
clearance from circulation. Moreover, during this period, there was no
detectable decrease in staining intensity for CD4 on PBLs, indicating that mAb
5A8 caused no significant CD4 modulation. A modest, transient decrease In CD4
cell number was noted in one monkey in coincidence with the appearance of the
anti-mouse immunoglobulin antibody response.

      To assess the effect that mAb 5A8 coating of CD4 cells might have on
immune function. T lymphocyte functional assays were performed on PBLs obtained
from monkeys receiving mAb. No consistent difference in responsiveness to PWM
(Fig. 5A) or an MLR (Fig. 5B) was observed between PBLs of monkeys given mAb 5A8
or control mAb. One anti-CD4-treated monkey did have a decreased MLR response on
day 14. However, this corresponded with the presence of an anti- mouse
immunoglobulin antibody response.

      The ability of the 5A8-treated monkeys to generate a humoral immune
response was also assessed. Monkeys were immunized with tetanus toxoid both
during and after mAb treatment As shown in Fig. 6, anti-tetanus immunoglobulin
responses were first detected in all monkeys by day 14 and were of similar
magnitude by day 21, despite continual CD4 cell coating with antibody for at
least 9 days. Similar results were obtained when tetanus-specific IgG was
measured (data not shown) Therefore, coating of CD4 lymphocytes with mAb 5A8
failed to induce any measurable abnormality in immune function.

                                   DISCUSSION

      Lymphocyte-specific monoclonal antibodies have been evaluated IN VIVO as
candidate immunosuppressive agents. The degree of immunosuppression induced in
animal models or humans has generally correlated with the ability of the mAb to
clear circulating cells or modulate functionally important molecules from cell
surfaces.(24) Moreover, the duration of immunosuppression has depended on the
length of time these effects could be maintained. In rodent models, rat
anti-mouse CD4 mAbs can completely and persistently clear circulating CD4
cells,(24-26) severely compromising the ability of the host to mount an immune
response. Other rat anti-mouse CD4 mAbs or mAb fragments that are nondepleting
but may modulate CD4 from the cell surface can also effectively inhibit immune
reactivity IN VIVO.(24,27) In addition, coadministration of depleting or
nondepleting anti-CD4 mAbs with a foreign antigen can induce a prolonged state
of antigen-specific tolerance, inhibiting the immune response to the mAbs
themselves as well as to the antigen.(24,26-29)

      CD4-specific mAbs have proved less predictable immunomodulators when
administered to nonhuman primates or to humans. Some mouse or mouse-human
chimeric antibodies have cleared peripheral blood CD4 cells(30-32) or induced
CD4 modulation from the surface of these cells(13,31,33) resulting in transient
suppression of immune responsiveness. Other anti-CD4 mAbs have caused no change
or minor transient decreases in PELs with varying degrees of immunosuppression.
(34-36) In the present study, treatment of normal monkeys with the mAb 5A8
neither cleared circulating CD4 cells nor induced CD4 modulation. The
lymphocytes of the monkeys retained their IN VITRO responsiveness to mitogens
and these monkeys generated normal humoral responses IN VIVO to an antigenic
challenge despite the coating of their CD4 cells with mAb. Thus, certain
CD4-specific
<PAGE>
mAbs can be given without inducing circulating cell clearance. CD4 modulation or
immunosuppression.

      The mAb used in these studies differs from most other CD4-specific mAbs
that have been previously evaluated IN VIVO in primates. Not only does mAb 5A8
not clear circulating CD4 cells or modulate the CD4 molecule from lymphocyte
surface, but administration ofmAb 5A8 resulted in an increase in circulating CD4
cells. These increases occurred within 1-3 days of the initial treatment. In
experiments not shown, mAb 5A8 did not affect the viability or growth IN VITRO
of normal or virus-infected PBL in 1 to 2-week cultures. The rapidity with which
mAb 5A8 increased the absolu1te number of circulating CD4 lymphocytes suggests
that this antibody may affect the trafficking of lymphocytes between circulating
and noncirculating pools. In addition to its high affinity for both human and
rhesus monkey CD4, this mAb was selected for these studies because it is a
po1tent inhibitor of an infection and cell fusion IN VITRO. This antibody also
has an unusual specificity, binding within domain 2 of CD4 and limiting
infection and fusion post virus binding.(21) We reasoned that this mAb might be
particularly useful as a blocking agent because it does not directly compete
with HIV for the CD4 domain 1 virus binding site, but rather inhibits
confonmational changes that may be required for HIV-induced fusion.(22)

      The appearance of an anti-mouse immunoglobulin humoral response limited
the effective duration of treatment with the mAb 5A8. A response of this type is
usually seen within 7-14 days of mouse mAb administration in nonhuman primates
or in humans.(37,38) Due to this response, mAb 5A8 was unable to coat
circulating CD4 cells when it was administered to a monkey 14 days after the
first treatment (data not shown). Thus, there was clear evidence for the
immunogenicity of this murine: protein in normal monkeys. However, preliminary
results have suggested that the humoral response to a foreign immunoglobulin may
be delayed or absent in monkeys with SIV((mac)-induced immunodeficiency.
Moreover, the antigenicity of a mAb might be minimized by utilizing mouse-human
chimeric antibodies or complementarity-determining region (CDR)-grafted forms.

      We have previously induced anti-CD4 autoantibodies by immunizing nonhuman
primates with recombinant forms of autologous or xenogeneic CD4.(14,15,39) This
antibody response correlated with an antiviral effect IN VITRO and IN VIVO. The
use of passive immunization would ensure both the appropriate antibody
specificity and uniform therapeutic levels of antibody. Such a form of therapy
could have utility as a postexposure prophylaxis in needle stick injuries, for
example. In addition, should maternal-fetal transmission of HIV occur during the
perinatal period, as in hepatitis B infection(40,41) then passive immunotherapy
with CD4-specific antibodies might effectively block this transmission.
Moreover, anti-CD4 antibodies could potentially decrease HIV replication in
chronically infected individuals.

      These studies demonstrate that anti-CD4 mAb administration can be
tolerated by normal monkeys and support the feasibility of using antibodies
directed against CD4 in HIV infection. Studies currently underway will evaluate
the ability of this mAb to inhibit replication IN VIVO and to block AIDS virus
infection using the rhesus monkey/SIV_animal model.

                                 ACKNOWLEDGMENT
<PAGE>
      The authors wish to thank Werner Meier, Konrad Miatkowski, and Renee
Shapiro for production and purification of the 5A8 and MOPC-21 antibodies used
in this study. We also thank Jennifer Blake, Dian Olson, and N. Mulrey for
excellent technical assistance and Shelley Kotlikoff for assistance in preparing
this manuscript. This work was supported by NIH Grants AI-20729. CA-51039, and
RR-000168 and by funds from Biogen, Inc. K.AR. is the recipient of a Special
Emphasis Research Career Award RR-OOOO55.

                                   REFERENCES

1.    Dalgleish AG, Beverley PCL, Chapman PR, Crawford DH, Greaves MF , and
      Weiss RA: The CD4(T4) antigen is an essential component of the receptor
      for the AIDS retrovirus. Nature (London) 1984;312:763-767.

2.    Klatzman D, Champagne E, Chamaret S, Gruest J, Guetard D, Hercend T,
      Gluckman J-C, and Montagnier L: T-lymphocyte T4 molecule behaves as the
      receptor for human retrovirus LAV.Nature (London) 1984;312:767-768.

3.    McDougal JS, Kennedy MS, Sligh IM, Con SP, Mawle A, and Nicholson JKA:
      Binding of HTLV-III/LAV to T4+T cells by complex of the 110K viral protein
      and the T4 molecule. Science 1986;231:382-385.

4.    Celada F, Cambiaggi C, Maccari J, Burastero S, Gregory T, Patzer E, Porter
      J, McDanal C, and Matthews T: Antibody raised against soluble CD4-pg120
      complex recognizes the CD4 moiety and blocks membrane fusion without
      inhibiting CD4-gp120 binding. I Exp Med 1990;172:1143-1150.

5.    Healey D, Dianda L, Moore JP, McDougal JS, Moore MJ, Estess P, Buck D,
      Kwong PD, Beverley PC, and Sartentau QJ: Novel anti-CD4 monoclonal
      antibodies separate human immunodeficiency virus infection and fusion of
      CD4+ cells from virus binding. J Exp Med 1990;172:1233-1242.

6.    Bank I and Chess L:Perturbation of the T4 molecule transmits a negative
      signal to T cells. J Exp Med 1985;162:1294-1303.

7.    Houghton AN and Larson S: Monoclonal antibodies for the treatment of
      cancer: Therapeutic strategies. Curr Opin Onco1 1989;1:258-265.

8.    Dillman, RO: Monoclonal antibodies for treating cancer. Ann Intern Med
      1989;111:592-603.

9.    Seimlan WE and Wofsy D: Selective manipulation of the immune repsonse IN
      VIVO by monoclonal antibodies. Annu Rev Med 1988;39:231-241.

10.   Dantal J and Soulillou JP: Use of monoclonal antibodies in human
      transplantation. Curr Opin Immunol 1991;3:740-747.

11.   Ziegler EJ, Fisher CJ Jr., Sprung CL, Straube RC, Sadoff JC, Foulke GE,
      Wortel CH, Fink MP, Dellinger RP , Teng NN, Allen IE, Berger HJ, Knatterud
      GL, LoBuglio AF, Smith CR, and the HA-1A Sepsis Study Group: Treatment of
      gram-negative bacteremia and septic shock with HA-1A human monoclonal
      antibody against endotoxin. New Engl J Med 1991;324:429-436.

12.   Greenman RL, Schein RM, Martin MA, Wenzel RP , MacIntyre NR, Emmanuel G,
      Chmel H, Kohler RB, McCarthy M, Plouffe J, Russell JA, and the XOMA Sepsis
      Study Group: A controlled clinical trial ofE5 murine monoclonal IgM
      antibody to endotoxin in the treatment of gram-negative sepsis. J Am Med
      Assoc 1991;266:1097-1102.
<PAGE>
13.   Dhiver C, Olive D, Rousseau S, Tamalet C, Lopez M, Galindo J-R, Mourens M,
      Hira M, Gastaut J-A, and Mawas C: Pilot phase I study using zidovudine in
      association with a 10-day course of anti-CD4 monoclonal antibody in seven
      AIDS patients. AIDS 1989;3:835-842.

14.   Waltanabe M, Chen ZW, Tsubota H, Lord CI, Levine CG, and Letvin NL:
      Soluble human CD4 elicits an antibody response is rhesus monkeys that
      inhibits simian immunodeficiency virus replication. Proc Natl Acad Sci USA
      1991;88:120-124.

15.   WaLtanabe M, Boyson JE, Lord CI, and Letvin NL: Recombinant soluble CD4
      immunized chimpanzees develop anti-self CD4 antibody responses with
      anti-HIV activity. Proc Natl Acad Sci USA 1992;89;5103-5107.

16.   Cassatt DR, Sweet RW, Arthos JA, and Truneh A: Immunization with soluble
      murine CD4 induces an anti-self antibody response without causing
      impairment of immune function. J Immunol 1991;147:1470-1476.

17.   Haynes BF, Dowell BL, Hensley LL, Gore I, and Metzgar RS: Human T cell
      antigen expression by primate T cells. Science 1982;215:298-300.

18.   Letvin NL, Todd III RF, Palley LS, Schlossman SF, and Griffin JD: T
      lymphocyte surface antigens in primates. Eur J Immunol 1983;13:345- 347.

19.   Desrosiers RC, Daniel MD, and Li Y:HIV-related lentiviruses of nonhuman
      primates. AIDS Res Hum Retroviruses 1989;5:465-473.

20.   Letvin NL and King NW: Immunologic and pathologic manifestations of the
      infection of rhesus monkeys with simian immunodeficiency virus of
      macaques. J Acquir Immune Defic Syndr 195)0;3:1023-1040.

21.   Burkly LC, Olson D, Shapiro R, Winkler G, Rosa JJ, Thomas DW, Williams C,
      and Chisholm P: Inhibition ofHIV infection by a novel CD4 domain
      2-specific monoclonal antibody. J Immunol 1992;149:1779-1787.

22.   Moore JP, Sattentau QJ, Klasse PJ, and Burkly LC: A monoclonal antibody to
      CD4 domain 2 blocks soluble CD4-induced conformational changes in the
      envelope glycoproteins of human immunodeficiency virus type I(HIV-1)
      and HIV-1 infection of CD4+ cells. J Virol 1992;66:4784-4793.

23.   Letvin NL, Chalifoux LV, Reimann KA, Ritz J, Schlossman SF, and Lambert
      JM: IN VIVO administration of lymphocyte-specific monoclonal antibodies in
      nonhuman primates. Delivery of ribosome-inactivating proteins to spleen
      and lymph node T cells. J Clin Invest 1986;78:666-673.

24.   Waldmann H: Manipulation of T-cell responses with monoclonal antibodies.
      Annu Rev Immuno1 1989;7:407-444.

25.   Alters SE, Sakai K, Steinman L, and Oi VT: Mechanisms of anti-CD4-mediated
      depletion and immunotherapy. A study using a set of chimeric anti-CD4
      antibodies. J Immunol 1990;144:4587-4592.

26.   Goronzy J, Weyand CM, and Fathman CG: Long-term humoral unresponsiveness
      IN VIVO induced by treatment with monoclonal antibody against L3T4. J Exp
      Med 1986;164:911-925.

27.   Carteron NL, Wofsy D, and Seaman WE: Induction of immune tolerance during
      administration of monoclonal antibody to L3T4 does not depend on depletion
      ofL3T4+ cells. J Immunol 1988;140:713-716.

28.   Shizuru JA, Taylor-Edwards C, Banks BA, Gregory AK, and Fathman CG:
      Immunotherapy of the nonobese diabetic mouse: Treatment with an antibody
      to T-helper lymphocytes. Science 1988;240:659-662.
<PAGE>
29.   Benjamin RJ and Waldman H: Induction of tolerance by monoclonal antibody
      therapy. Nature (London) 1986;320:449-451.

30.   Horneff G, Burmester GR, Ernmrich F, and Kalden JR: Treatment of
      rheumatoid arthritis with an anti-CD4 monoclonal antibody. Arthritis Rheum
      1991;34:129-140.

31.   Jonker M, Neuhaus P, Zurcher C, Fucello A, and Goldstein G: OKT4 and OKT4A
      antibody treatment as immunosuppression for kidney transplantation in
      rhesus monkeys. Transplantation 1985;39:247-253.

32.   Prinz J, Braun-Falco 0, Meurer M, Daddona P, Reiter C, Reiber P, and
      Riethmu1ler G: Chimaeric CD4 monoclonal antibody in treatment of
      generalized pustular psoriasis. Lancet 1991;338:320-321.

33.   Cosimi AB, Delmonic FL, Wright JK, Wee SL, Preffer FI, Jolliffe LK, and
      Colvin RB: Prolonged survival of non-human primate renal allograft
      recipients treated only with anti-CD4 monoclonal antibody. Surgery
      1990;108:406-413.

34.   Reiter C, Kakavand B, Rieber EP, Schattenkirchner M, Riethmuller G, and
      Kruger K: Treatment of rheumatoid arthritis with monoclonal CD4 antibody
      M-T151. Arthritis Rheum 1991;34:525-536.

35.   Hafter DA, Ritz I, Schlossman SF, and Weiner HL: Anti-CD4 and Anti-CD2
      monoclonal antilbody infusions in subjects with multiple sclerosis. J
      lmmunol 1988;141:131-138.

36.   Rose LM, Alvord JR EC. Hruby S, Jackevicius S, Petersen R, Warner N. and
      Clark EA: IN VIVO administration of anti-CD4 monoclonal antibody prolongs
      survival in longtailed macaques with experimental allergic
      encephalomyelitis. Clin Immunol lmmunopathol 1987;45:405-423.

37.   Reimann KA, Turner S, Lambert JM, Reed MH, Schlossman SF, and Letvin NL:
      IN VIVO administration of lymphocyte-specific monoclonal antibodies in
      nonhuman primates. V. Evidence that humoral immune response to monoclonal
      antibodies and immunotoxin conjugates abrogates their cytotoxic activity.
      Transplantation 1989;48:906-912.

38.   Jaffers GJ, Fuller TC, Cosimi AB, Russell PS, Winn HJ, and Colvin RB:
      Monoclonal antibody therapy: Anti-idiotypic and non-anti-idiotypic
      antibodies to OKT3 arising despite intense immunosuppression.
      Transplantation 1986;41:572-578.

39.   Watanabe M, Levine CG, Shen L, Fisher RA, and Letvin NL: Immunization of
      simian immunodeficiency virus-infected rhesus monkeys with soluble human
      CD4 elicits an antiviral response. Proc Natl Acad Sci USA
      1991;88:4616-4630.

40.   Yeoh EK: Hepatitis B virus infection in children. Vaccine
      1990;8(Suppl):S29-S30.

41.   Wright R: Viral hepatitis: Comparative epidemiology. Br Med Bull
      1990;46:548-558.

                                                    Address reprint requests to:
                                                                KEITH A. REIMANN
                                    NEW ENGLAND REGIONAL PRIMATE RESEARCH CENTER
                                                          HARVARD MEDICAL SCHOOL
                                                                   P.O. BOX 9102
                                          SOUTHBOROUGH, MASSACHUSETTS 01772-9102


*New England Regional Primate Research Center, Harvard Medical School, One Pine
Hill Drive- .Southborough, Massachusetts 01772.
(T)BiogEn, Inc., 14 Cambridge Center, Cambridge, Massachusetts 02142.

<PAGE>
FIG. 1. Monoclonal antibody 5A8 has similar affinity for human and Thesus monkey
CD4(+) PBLs. Peripheral blood lymphocytes from normal humans or rhesus monkeys
were stained with varying concentrations of 5A8 by an indirect technique and
analyzed by flow cytometry. Mean fluorescence +/- SD, (n)=2.(o) Human PBLs; (.)
rhesus monkey PBLs.

FIG. 2. A single intravenous administration of anti-CD4 monoclonal antibody 5A8
coats circulating CD4 lymphocytes without causing CD4(+) cell clearance. Two
monkeys received 3 mg of 5A8 per kilogram intravenously on day 0 (arrow). Data
points from day 0 represent samples collected prior to treatment. (A) Serum
levels of mouse immunoglobulin (micrograms/ml): (B) absolute number of CD4(+)
PBLs per microliter: (C) percentage ofPBLs that were CD4(+); (D) percentage of
CD4(+) PBLs that were coated with 5A8; (E) monkey anti-mouse immunoglobulin
titer.

FIG. 3. A single intravenous administration of anti-CD4 monoclonal antibody 5A8
coats all CD4(+) cells within peripheral lymph nodes. Peripheral lymph node
biopsies were obtained on the indicated days following a single 3-mg/kg
intravenous administration of 5A8 and assessed for the presence of surface-bound
5A8 by staining with FITC-conjugated goat anti-mouse immunoglobulin and analyzed
by flow cytometry(left). To determine the total number ofCD4(+) cells, samples
were also incubated with additional 5A8 and then stained with FITC-conjugated
goat anti-mouse immunoglobulin(right). The background fluorescence in each
sample is indicated by the unfilled tracing.

FIG. 4. Repeated administration of anti-CD4 monoclonal antibody 5A8 resulted in
continuous coating of CD4 lymphocytes for 9 days without clearing cells from
circulation. Two monkeys were intravenously administered 3 mg of 5A8 (.) per
kilogram or a control antibody (...) on days 0, 2, and 4 (treatments indicated
by arrows). (A) Percentage of CD4(+) cells in PBLs. (B) Absolute number of
CD4(+) cells per microliter. Mean +/- SD. Shaded bar represents time during
which CD4 cells were coated with mAb.

FIG. 5. Peripheral blood lymphocytes from monkeys receiving three intravenous
administrations of anti-CD4 monoclonal antibody had normal IN VITRO
proliferative responses. Peripheral blood lymphocytes were obtained on the
indicated days from monkeys receiving 5A8 (black bars) or a control antibody
(shaded bars), cultured with PWM(A) or in a xenogeneic MLR(B) and IN VITRO
proliferation was assessed. An experimental and control monkey were paired for
each experiment.

FIG. 6. Monkeys receiving three intravenous administrations of anti-CD4
monoclonal antibody elicit normal humoral responses to primary immunization with
tetanus toxoid. Monkeys that received three administrations of 5A8 or control
antibody (treatments indicated by lower arrows) were immunized with tetanus
toxoid on day 3 and day 21 (TT). Anti-tetanus serum immunoglobulin liters were
assessed as indicated. (o) 5A8-treated monkeys; (") control-treated monkeys.
Shaded bar represents time during which CD4 cells were coated with mAb.

TABLE I. ANTI-CD4 mAb 5A8 SUPPRESSES HIV-1 AND SIV(mac) REPLICATION.

                                          HIV p24 or SIV p27 (ng/ml)
<PAGE>


                                       In vitro infection of human                               In vitro infection of H9
                                            PBLs with HIV(MN)                                       cells with SIV(mac)

                                                                                                         
Antibody               Day 3(a)                Day 7             Day 10             Day 3              Day 7               Day 10
added to
cultures
No mAb             greater than 0.5(b) greater than 0.5   greater than 0.5   greater than 1.0   greater than 1.0   greater than 1.0
Irrelevant mAb            NT(c)                NT                 NT         greater than 1.0   greater than 1.0   greater than 1.0
mAb 5A8
0.1 micrograms/ml     less than 0.5           0.1          less than 0.5            0.5         greater than 1.0   greater than 1.0
1.0 micrograms/ml     less than 0.5      less than 0.05    less than 0.05           0.6         greater than 1.0    less than 0.1
10.0 micrograms/ml    less than 0.5      less than 0.05    less than 0.05      less than 0.1      less than 0.1     less than 0.1


                            CD8-depleted PBLs from                                          CD8-depleted PBLs from
                             HIV-1-infected human                                           SIV(mac)-infected monkey

                          Day 3              Day 8                                  Day 3              Day 5           Day 10
No mAb              greater than 0.5   greater than 0.5                      greater than 1.0   greater than 1.0   greater than 1.0
Irrelevant mAb      greater than 0.5   greater than 0.5                             NT                 NT                NT
mAb 5A8
0.1 micrograms/ml         0.47         greater than 0.5                      greater than 1.0          0.6               0.3
1.0 micrograms/ml         0.08          less than 0.05                              1.0                0.3               0.1
10.0 micrograms/ml        0.12          less than 0.05                       greater than 1.0          0.3               0.2


(a) Cultures were established as described in Materials and Methods. Culture
    medium was assayed, with a commercial kit, on the indicated days for HIV p24
    or SIV p27.

(b) Sensitivity of assay was 0.05 ng/ml for HIV p24 and 0.1 ng/ml for SIV p27.

(c) NT, Not tested.

<PAGE>
TABLE 2. CHANGE IN ABSOLUTE PERIPHERAL BLOOD LYMPHOCYTE NUMBER FOLLOWING
         ANTI-CD4 ADMINISTRATION


                    Cells/microliters (percentage change)(a)

                          Lymphocytes                         CD4(+)lymphocytes

                                                      
Treatment(b)  Day 0  Day 2(c)     Day 4       Day 6     Day 0   Day 2      Day 4
Single        1634   2546(+56)   2380(+46)   1540(-6)     801   1375(+72)  1214(+52)
Single        3060   6902(+125)  3119(+2)    1449(-53)   1102   3037(+176) 1560(+42)
Multiple      1836   3135(+71)   2938(+60)   6976(+280)   973   1850(+90)  1704(+75)
Multiple      2580   2745(+6)    5166(+100)  2624(+2)    1290   1702(+32)  3152(+144)


Percentage change was calculated by dividing the posttreatment value by the
pretreatment value.
Monkeys received either a single 3-mg/kg intravenous administration of anti-CD4
mAb, or three administrations (multiple) of 3
on days 0,2, and 4. In multiple infusion experiments, blood samples for analysis
were drawn prior to treatment.
Values in bold-face type represent time points when CD4 cells were coated with
monoclonal antibody.


<PAGE>
AIDS RESEARCH AND HUMAN RETROVIRUSES
Volume 11, Number 4, 1995
Mary Ann Liebert, Inc., Publishers

      IN VIVO ADMINISTRATION OF CD4-SPECIFIC MONOCLONAL ANTIBODY: EFFECT ON
      PROVIRUS LOAD IN RHESUS MONKEYS CHRONICALLY INFECTED WITH THE SIMIAN
                       IMMUNODEFICIENCY VIRUS OF MACAQUES
  KEITH A. REIMANN(1), RICHARD L. CATE(2), YAMING WU(2), LOUISE PALMER(2), DIAN
   OLSON(2), BARRY C.D. WAITE(1), NORMAN L. LETVIN(1), and LINDA C. BURKLY(2)

ABSTRACT

Since monoclonal antibodies (MAb) specific for CD4 are potent inhibitors of HIV
and SIV replication IN VITRO, we explored their potential usefulness IN VIVO as
an AIDS therapy. The anti-CD4 MAb 5A8 binds to domain 2 of the CD4 molecule and
inhibits virus replication and virus-induced cell fusion at a postvirus binding
step. Administration of this MAb to normal rhesus monkeys coats all circulating
and lymph node CD4 cells and induces neither CD4 cell clearance nor measurable
immunosuppression. In the present study, monkeys chronically infected with the
simian immunodeficiency virus of macaques (SIVmac) had stable levels of SIVmac
provirus in PBMC prior to treatment as measured by a quantitative polymerase
chain reaction technique. Six infected monkeys treated with anti-CD4 MAb
demonstrated a significant decrease in SIVmac provirus level after 9 days. Of
these monkeys, 3 had >800 CD4 cells/Ml and developed strong antimouse Ig
responses that prevented further treatment. The remaining 3 monkeys had <800 CD4
cell/Ml and failed to develop antimouse Ig antibody responses. When treatment
was continued for 12-21 days in these monkeys, a sustained or further decrease
in SIVmac provirus load occurred over the extended treatment period. Four
monkeys that received a control MAb of irrelevant specificity for 9-22 days
showed either no significant change or a transient increase in SIVmac provirus.
Thus, the passive administration of anti-CD4 MAb may exert a specific antiviral
effect in controlling immunodeficiency virus infection IN VIVO.


                                  INTRODUCTION

The CD4 molecule that acts as a high-affinity receptor for the envelope
glycoprotein of HIV is a potential target for therapeutic intervention in HIV
infection.1-4 Indeed, certain anti-CD4 monoclonal antibodies (MAbs) can
efficiently block HIV-induced syncytia formation and infection of lymphocytes
and macrophages(1,2,5) or spread of infection among CD4+ target cells IN
VITRO.(6,7) However, this mode of passive immunotherapy has received little
attention.(6-9)

      The simian immunodeficiency virus of macaques (SIVmac) is a lentivirus
similar to HIV both genetically and in its CD4 cell tropism and induces an
AIDS-like disease in rhesus monkeys(10,11). Additionally, many monoclonal
antibodies directed against the human CD4 molecule also react with rhesus monkey
CD4 due to the phylogenetic proximity of man and the rhesus monkey(12,13)


<PAGE>
These features make this animal model uniquely suited to evaluate CD4-directed
therapies for AIDS.

      Recently, PCR technology has been employed to quantitate the HIV provirus
levels in HIV-infected individuals offering a powerful alternative to
quantitation of viral burden by serum antigenemia or virus culture from
blood.14,15 Although a clear correlation between DNA provirus and antiretroviral
therapy is not yet established, some have reported decreases in HIV provirus in
response to nucleoside analogs.16-19 By analogy, we sought to apply quantitative
DNA PCR to assess the SIV proviral burden in chronically SIV-infected rhesus
monkeys.

      Previously we have explored the safety of IN VIVO administration of
CD4-specific MAbs in normal rhesus monkeys.20 This anti-CD4 MAb 5A8 is a potent
inhibitor of HIV and SIV infection and cell fusion IN VITRO20,21 and appears to
inhibit HIV infection/fusion in a noncompetitive fashion, blocking postvirus/CD4
binding steps of infection.21,22 When administered intravenously to normal
rhesus monkeys, this MAb bound to all CD4+ lymphocytes in the peripheral blood
and lymphoid organs. Despite coating with MAb, CD4 cells were not cleared from
circulation, but rather increased transiently in number. Neither did this
antibody induce any measurable immunodeficiency. However, the duration of
treatment was limited by the appearance of an antimouse Ig antibody
response.(20)

      In the present study, we explored the feasibility of using anti-CD4 MAb as
passive immunotherapy in the treatment of AIDS employing rhesus monkeys
chronically infected with SIVmac. We show that chronic administration of a
CD4-specific MAb results in 5- to 10-fold decreases in virus load as measured by
changes in SIV provirus.


                              MATERIALS AND METHODS

PREPARATION OF MONOCLONAL ANTIBODIES

      The anti-CD4 MAb 5A8 was isolated and characterized as described
previously.(21,22) 5A8 and isotype-matched control MAb, MOPC-21, were purified
by protein A affinity chromatography from tissue culture supernatants. Purified
MAbs contained less than 3 endotoxin units/mg.

ANIMALS AND ANTIBODY ADMINISTRATION

      The rhesus monkeys (MACACA MULATTA) used in this study were maintained in
accordance with the guidelines of the Committee on Animals for the Harvard
Medical School and the Guide for the Care and Use of Laboratory Animals
[Department of Health and Human Services Publication No. (NIH) 85-23, revised
1985]. All monkeys were infected by intravenous inoculation with uncloned simian
immunodeficiency virus of macaques (SIVmac), strain 251, which was propagated in
human PBMC. Monkeys were anesthetized with ketamine-HCl for all procedures.

      Monkeys were administered 3 mg/kg of MAb 5A8 or a control antibody diluted
to 1-2 mg/ml in phosphate-buffered saline (PBS) by intravenous injection over
approximately 1 min. Antibody was

<PAGE>
administered every 3 days until monkeys developed a significant antimouse Ig
titer. The minimum number of injections given was 4 and the maximum was 8. The
eighth injection was given on day 22 rather than day 21.

IMMUNOPHENOTYPING OF MONKEY PBL

      Unfractionated cells from EDTA anticoagulated blood were washed three
times with PBS/0.1% bovine serum albumin (BSA) to remove plasma and previously
administered MAb. Then, cells were resuspended in PBS/0.1% BSA at the original
blood volume and immunophenotyped as previously described.23 Briefly, 50 Ml of
blood cell suspension was added to 100 Ml of PBS/0.1% BSA that contained an
appropriate amount of MAb. MAbs used were anti-CD4 (OKT4-FITC; Ortho Diagnostic
Systems, Raritan, NJ) and anti-CD45 (LCA, Dako Corp., Carpenteria, CA) Cells
were incubated for 30 min at room temperature and then washed once with PBS. In
the stains for CD45, an indirect staining technique was used. Cells were
incubated, additionally, with a 1:50 dilution of FITC-conjugated goat antimouse
Ig [F(ab)2 fragment] (Jackson Immuno-Research, West Grove, PA) in PBS/0.1% BSA
for 20 min at room temperature and then washed with PBS. RBC in all samples were
lysed using a commercial RBC lysing kit (Whole Blood Lysing Reagent Kit; Coulter
Corp., Hialeah, FL), washed in PBS, and resuspended in 0.4 ml of PBS/1%
formalin. Samples were analyzed routinely on a flow cytometer using forward
light scatter, 90(degree) light scatter, and CD45 fluorescence to identify
lymphocytes.

ASSAY FOR ANTIMOUSE IG LEVELS

      Titer of antimouse Ig in monkey plasma was determined as described
previously using an ELISA in which 5A8 was adsorbed to microtiter plates and
binding monkey Ig was detected using an HRP-conjugated antihuman Ig antiserum
that cross reacts with rhesus monkey Ig.20 Antimouse Ig levels were determined
prior to treatment and following treatment on the days indicated.

QUANTITATION OF IN VIVO SIVMAC PROVIRUS BY PCR

      PBMC isolated on Ficoll gradient were washed with PBS and frozen as cell
pellets (1-3 x 106 cells) in liquid nitrogen. Samples were thawed and cell
pellets were disrupted after adding 0.4 ml of lysis buffer (0.5 M EDTA, pH 8.0,
0.5% Sarkoysyl, 100 Mg/ml proteinase K). Lysates were incubated for 4 hr at
50(degree)C, and extracted two times with equal volumes of phenol/chloroform
(1:1). DNA was precipitated by adding 0.9 volumes of room temperature ethanol
followed by centrifugation at room temperature for 2 min in an Eppendorf
centrifuge. The DNA pellets were resuspended in TE (10 mM Tris, pH 8.0, 1 mM
EDTA) and precipitated a second time with 3 volumes of ethanol, followed by
resuspension in 100 Ml TE. The total cellular DNA was quantitated on a 0.7%
agarose gel.

      Two sets of PCR primers were used to quantitate the level of SIVmac
provirus in the DNA samples:

      SIV1 5_GACCATGTTATGGCCAAATG3_and

<PAGE>
      SIV2 5_GGCTTCTCTCTGCTTTCTCT3_;

      RCD1 5_TTTCCAGAAGGCCTCCAGCA3_and

      RCD2 5_AGGTTCACTTCCTGATGCAA_3.

Primers SIV1 and SIV2 are from the region of the SIVmac-251 viral genome where
the POL and GAG gene products overlap and produce a 200 base pair PCR product.
Primers RCD1 and RCD2 produce a 333 base pair PCR product from the rhesus monkey
CD4 gene, which serves as a reference gene present in DNA derived from PBMC. PCR
was performed at a final concentration of 1 x PCR buffer (10 mM Tris-HCl, pH
8.3, 50 mM KCl, 1.5 mM MgCl2, 0.01% gelatin), 0.2 mM dNTPs, 0.25 MM primers
(SIV1, SIV2, RCD1, RCD2), 2.5 units of THERMUS AQUATICUS DNA polymerase (TAQ
polymerase, Perkins-Elmer/Cetus, Norwalk, CT), and 200 ng genomic DNA in a final
volume of 100 Ml. Genomic DNA and primers were heat denatured for 2 min and
quick-chilled on ice prior to setting up the PCR reaction. The mixture was
overlaid with mineral oil and amplified 24 cycles in a thermal cycler
(Perkin-Elmer/Cetus). The amplification profile involved denaturation at
93(degree)C for 1 min. primer annealing at 50(degree)C for 1 min. and extension
at 72(degree)C for 1 min. On the last cycle, the extension time was increased to
8 min. Twenty microliters of each PCR reaction was electrophoresed on 2% agarose
gels in Tris- acetate/EDTA buffer. The DNA was transferred to nylon membranes
(Gene Screen; Dupont, Boston, MA), and fixed by UV crosslinking.(24)

      The nylon membranes were hybridized with two oligomers, SIV3, which
hybridized to the 200 base pair SIV PCR fragment, and RCD3, which hybridized to
the 333 base pair CD4 PCR fragment:

      SIV3 5_TCTTTAGCAGATCCACAGCTGGGT3_

      RCD3 5_GAGCTTCTTGCCCATCTGGAGCTT3_

The two oligomers were labeled with [y-32P]ATP (6000 Ci/mmol; Dupont) using
polynucleotide kinase and separated from unincorporated nucleotides by
electrophoresis on a 20% acrylamide gel.(25) The membranes were prehybridized 30
min at 50(degree)C in 1% crystalline grade bovine serum albumin, 0.5 M sodium
phosphate, pH 7.2, 7% SDS, 1 mM EDTA.(26) Hybridization was performed at
50(degree)C for 12 hr in 100 ml of the same buffer containing 20 pmol of the
32P-labeled SIV3 oligomer, 10-20 fmol of the 32P-labeled RCD3 oligomer, and 20
pmol of the unlabeled RCD3 oligomer. PCR products were detected with the
SIV-specific oligomer and 1000- to 2000-fold less 32P-labeled CD4-specific
oligomer in order to avoid overexposure of the CD4 bands when detecting the SIV
PCR products. After hybridization, the membranes were washed four times for 5
min at 50(degree)C with 5 x SSC/1% SDS and two times for 5 min with 3.2 M
tetramethylammonium chloride (TMAC1)/1% SDS at 55(degree)C. Membranes were
exposed to Kodak XAR-5 film for 1-4 days at -70(degree)C using intensifying
screens. Band intensity was determined using a densitometer (Molecular Dynamics
Model 300A, Sunnyvale, CA). The provirus level in PBMC was calculated using the
ratio of SIV gene PCR product/CD4 gene PCR product expressed as a percentage
after correcting for the specific activity of the CD4 probe.

<PAGE>
      To determine the precision and reproducibility of this assay, both the
intraassay and interassay variation were measured. In 74 samples run as
replicates within the same assay, the median coefficient of variation was 21%
with a range of 5-51%. In 50 samples run twice in independent assays, the median
coefficient of variation was 22% with a range of 3-76%. Thus, the variation was
similar whether replicates were run in a single assay or if the sample was
analyzed in independent assays. Further, these results indicate that the assay
can detect differences in SIV provirus levels greater than 2-fold.

STATISTICAL ANALYSIS

      Each value used as an estimate of SIV provirus was obtained as a mean of 2
assays performed on the same blood sample. Pretreatment mean values were derived
from duplicate assays that were performed on 2 independent blood sampling time
points. Differences in SIVmac provirus levels between control and experimental
groups were determined using ANOVA (balanced design, one group factor, one
repeated measures factor). Pairwise comparisons between posttreatment values and
the pretreatment mean value were made using Neuman-Keul's multiple range test.
For all analyses, P < 0.05 was considered significant.



                                     RESULTS

ADMINISTRATION OF MAb TO SIVMAC-INFECTED MONKEYS

      Ten rhesus monkeys that were chronically infected with SIVmac were
administered the anti-CD4 MAb (n = 6), or an isotype-matched control MAb (n = 4)
every third day by intravenous injection. In our previous studies using this
antibody in normal rhesus monkeys, a humoral immune response to mouse Ig
appeared in the serum between 9 and 12 days after the first administration.20
Further administration of 5A8 to monkeys with high antimouse Ig titers resulted
in a failure of 5A8 to bind to CD4+ lymphocytes and in an acute systemic
allergic reaction. Therefore, in the present study, we measured the antimouse Ig
titer prior to each treatment after day 9 to determine whether additional
treatments could be performed. Four of ten monkeys rapidly developed high-titer
antimouse Ig responses after 4 MAb injections by day 12 and further treatments
could not be administered to these animals (Fig. 1). In the 6 remaining monkeys,
antimouse Ig responses remained orders of magnitude lower. Thus, treatments were
continued in 4 monkeys for a total of 5-8 injections (Table 1). When circulating
CD4 counts were examined, all 4 monkeys with rapidly appearing, antimouse Ig
responses had CD4 counts greater than 800/microliters. In contrast, only 1 of 6
monkeys with low or delayed antimouse Ig responses had a CD4 count
>800/microliters. Therefore, monkeys with lower CD4 cell counts and presumably
more marked immune dysfunction could be treated longer with mAb.

CHANGES IN PBMC IN RESPONSE TO TREATMENT WITH MAb

      We have shown previously that treatment of normal rhesus monkeys with MAb
5A8 results in a transient increase in circulating CD4 cell number.(20) This
increase appears very soon after treatment suggesting that it results from
changes in trafficking of CD4+ lymphocytes rather than from an

<PAGE>
actual increase in total CD4 cell number. A similar transient increase in
circulating CD4 cell number was noted in 4 of 6 SIVmac-infected monkeys treated
with 5A8 (Fig. 2) while no change in PBMC phenotype was noted in monkeys treated
with the irrelevant MAb (data not shown). In no case was a sustained decrease in
CD4+ cells observed. In addition, changes in monocyte number did not correlate
with MAb treatment (Table 1).

QUANTITATIVE PCR FOR DETECTION OF SIV PROVIRUS LOAD

   We sought to measure changes that occurred in the quantity of provirus in
PBMC during the MAb treatment of this cohort of chronically SIVmac-infected
rhesus monkeys. Serum antigenemia and virus isolation by cocultivation of PBMC
from infected monkeys with SIVmac-permissive cells were not consistently
detectable. Thus, a sensitive, quantitative PCR assay was employed. DNA from
chronically SIVmac-infected H9 cells was diluted into DNA from uninfected
rhesus monkey PBMC at varying ratios (1:250, 1:500, 1:1000, 1:2000) for use as
an assay standard. Figure 3A shows the level of SIV proviral DNA and that of a
reference gene (rhesus CD4) after PCR amplification of these samples.
Densitometric quantitation allowed the ratio of the SIV/CD4 PCR products to be
calculated and these values are plotted for each input DNA sample as shown in
Figure 3B. SIV/CD4 PCR products of equal intensity were expected at a 1/2000
ratio of SIV/CD4 input DNA since the specific activity of the CD4 probe was
2000-fold less than the SIV probe. Thus, these data demonstrate that the ratio
of the PCR products reflects the SIV/CD4 ratio in the corresponding input DNA
and that the PCR amplification performed under these conditions is reproducible,
linear and sensitive to 2-fold differences. The coefficient of variation in this
assay ranged from 8 to 35% and thus was similar to the variation observed in
assays of PBMC from infected monkeys.

PROLONGED 5A8 MAB TREATMENT OF CHRONICALLY SIV-INFECTED MONKEYS DECREASES SIV
MAC PROVIRUS LEVELS

   SIVmac provirus levels were evaluated by quantitative PCR in SIV
mac-infected monkeys treated either with 5A8 or an isotype-matched control Ig.
SIVmac provirus was consistently detected in the PBMC of all 10 infected
monkeys studied. Figure 4 illustrates the autoradiographs obtained for provirus
quantitation from 5 representative monkeys. The relative SIV provirus levels for
all monkeys are shown in Figure 5.

   Provirus was quantitated at 2 or 3 times in each monkey before treatment.
While variation in provirus level occurred between monkeys, these pretreatment
values varied less than 2-fold on average within each individual. Statistical
analysis confirmed that there were no differences between replicate samples
obtained at the time points before treatment or between mean pretreatment values
in the control vs experimental groups.

   The 6 monkeys that received the MAb 5A8 every 3 days for 9 days exhibited a
significant decrease in SIV provirus levels as compared to the 4 control treated
monkeys (ANOVA, balanced design, one group factor, one repeated measures factor)
(Fig. 6). Pairwise comparisons between pre- and posttreatment means showed
significantly lower SIV provirus level in experimental animals on day 12
(Neuman-Keuls multiple range test). Power calculations on these data indicated
that

<PAGE>
increasing the experimental n by 2- or 3-fold would result in statistically
significant differences between pre- and posttreatment values and between
experimental and control groups at 9 and 15 days as well.(27)

   Three of these 6 5A8-treated monkeys failed to develop a high-titer antimouse
Ig response and these monkeys were treated for a total of 12 to 22 days. All
showed sustained or further reductions in SIV provirus level over this extended
treatment period. Monkey number 163 exhibited a 5-fold decrease through day 15
and numbers 81 and 152 both achieved 10-fold reductions through day 30 (Fig.
5C). These decreases in provirus level in the CD4-depleted group persisted for
at least 5-7 days after the last MAb administration. In all monkeys, provirus
levels tended to return to pretreatment values after treatment was discontinued.

   A similar decrease in SIV provirus was not observed in monkeys treated with a
control MAb. Rather control monkeys exhibited no significant change in SIV
provirus level whether treated short term (9 days, monkeys 239, 107, 410) or
long term (22 days, monkey 177) (Figs. 5A and 6). In fact, the increased
variation in mean provirus levels on day 6 and day 9 reflects a transient
increase in SIV provirus in monkey 177. Thus, provirus levels were markedly and
reproducibly reduced relative to pretreatment levels in these 5A8-treated
monkeys, whereas the corresponding control Ig-treated animals exhibited no
significant change.

                                   DISCUSSION

   We have shown that monoclonal antibodies specific for CD4 can be administered
to SIVmac-infected rhesus monkeys without clearing of circulating CD4 cells.
This murine MAb could be administered for over 3 weeks in some animals when an
AIDS virus-induced immunodeficiency modulated the immune response to the mouse
Ig. To assess the efficacy of this treatment in controlling SIVmac replication,
we developed a quantitative assay to measure SIVmac provirus. SIVmac DNA was
readily detected and generally varied only 2-fold within individuals in repeated
samplings obtained prior to treatment. Surprisingly, one monkey (177) that
received an irrelevant MAb developed an increase in SIVmac provirus in the PBMC.
The paradoxical increase in SIVmac provirus seen in the monkey treated with an
irrelevant MAb may simply be due to increased virus replication as a result of
immune cell activation by treatment with foreign protein. Activation of infected
CD4 cells increases virus replication and may serve to facilitate infection of
other CD4 cells in an individual(28). Conversely, when an anti-CD4 MAb was
administered, SIVmac provirus levels decreased. Most notably, SIVmac-infected
animals that were more severely compromised (CD4 counts <800/microliters) and
could be treated for a longer period of time had a marked and sustained decrease
in measured proviral load. This decrease in SIVmac provirus appeared to be
specific since treatment of control monkeys with a similar disease profile for
the same duration did not show a proviral reduction.

   This decrease in SIV provirus in PBMC was not simply due to clearance of CD4+
lymphocytes from the circulation. Lymphocytes from normal monkeys that bound 5A8
IN VIVO were not cleared from the circulation but, in fact, increased in number.
This same transient increase in circulating CD4 cells was seen in some
SIV-infected monkeys treated with 5A8. Since in our provirus assay

<PAGE>
we did not adjust for this increased percent CD4+ cells, we may have actually
underestimated the decrease in provirus at the early time points.

   Our method of quantitating virus load measures provirus only in cells in the
peripheral blood compartment. There is mounting evidence that virus burden may
vary in the different lymphoid compartments.(29,30) Although there was no
measurable decrease in circulating CD4+ cell number, we cannot rule out that the
decrease in SIVmac provirus within the PBMC compartment following 5A8 treatment
was due to changes in recirculation of CD4 cells caused by coating with MAb.
However, for changes in cell trafficking alone to have this effect, the MAb-
induced change would have to selectively recruit uninfected CD4 cells into the
circulation.

   Humans infected with HIV might respond to treatment with anti-CD4 MAb in a
similar way. This antibody was shown to block IN VITRO replication of SIV and
HIV with equivalent efficiency.(20) However, the deleterious effect of
administering mouse Ig and the duration of treatment may be improved
substantially by using CDR-grafted, "humanized" MAb. In addition to treating
chronically HIV-infected individuals, this form of passive immunotherapy may
have usefulness in blocking primary infection with HIV in the setting of
needle-stick injuries or other occupational exposures(6,7). These results
suggest that CD4-directed treatments may exert a potent antiviral effect in
individuals infected with an AIDS virus.

                                ACKNOWLEDGMENTS

   We thank Alan Gill and Bernard J. Ransil for valuable discussions. Data
organization and analysis were performed on the PROPHET system, a national
computer resource sponsored by the National Center for Research Resources,
National Institutes of Health. This work was supported by National Institutes of
Health Grants RR-00168 and CA-50139. Special Emphasis Research Career Award
RR-00055 (K.A.R.), and by funds from Biogen, Inc.

                                   REFERENCES

1.  Dalgleish AG, Beverley PCL, Chapham PR, Crawford DH, Greaves MF, and Weiss
    RA: The CD4(T4) antigen is an essential component of the receptor for the
    AIDS retrovirus. Nature (London) 1984;312:763-767.

2.  Klatzman D, Champagne E, Chamaret S, Gruest J, Guetard D, Hercend T,
    Gluckman J-C, and Montagnier L: T-lymphocytes T4 molecule behaves as the
    receptor for human retrovirus LAV. Nature (London) 1984;312:767-768.

3.  McDougal JS, Kennedy MS, Sligh JM, Cort SP, Mawle A, and Nicholson JKA:
    Binding of HTLV-III/LAV to T4+ T cells by complex of the 110K viral protein
    and the T4 molecule. Science 1986;231:382-385.

4.  Sattentau QJ and Weiss RA: The CD4 antigen: Physiological ligand and HIV
    receptor. Cell 1988;52:631-633.

<PAGE>
5.  McDougal JS, Mawle A, Cort SP, Nicholson JKA, Cross GD, Scheppler-Campbell
    JA, Hicks D, and Sligh J: Cellular tropism of the human retrovirus
    HTLV-III/LAV. I. Role of T cell activation and expression of the T4 antigen.
    J Immunol 1985;135:3151-3162.

6.  Attanasio R, Allan JS, and Kennedy RC: Monoclonal anti-CD4 as
    immunoprophylactic agents for human immunodeficiency virus infection. J
    Infect Dis 1993;168:515-516.

7.  Rieber EP, Federle C, Reiter C, Krauss S, Gurtler L, Eberle I, Deinhardt F,
    and Riethmuller G: The monoclonal CD4 antibody M-T413 inhibits cellular
    infection with human immunodeficiency virus after viral attachment to the
    cell membrane: An approach to postexposure prophylaxis. Proc Natl Acad Sci
    USA 1992;89:10792-10796.

8.  Dhiver C, Olive D, Rousseau S, Tamalet C, Lopez M, Galindo J-R, Mourens M,
    M Hira M, Gastaut J-A, and Mawas C: Pilot phase I study using zidovudine in
    association with a 10-day course of anti-CD4 monoclonal antibody in seven
    AIDS patients. AIDS 1989;3:835-842.

9.  Reiber EP, Reiter C, Gurtler L, Deinhardt F, and Reithmuller G: Monoclonal
    CD4 antibodies after accidental HIV infection. Lancet 1990;336: 1007-1008.

10. Desrosiers RC, Daniel MD, Li Y: HIV-related lentiviruses of non-human
    primates. AIDS Res Hum Retroviruses 1989;5:465-473.

11. Letvin NL and King NW: Immunologic and pathologic manifestations of the
    infection of rhesus monkeys with simian immunodeficiency virus of macaques.
    J Acquir Immune Defic Syndr 1990;3:1023-1040.

12. Haynes BF, Dowell BL, Hensley LL, Gore I, and Metzgar RS: Human T cell
    antigen expression by primate T cells. Science 1982;215:298-300.

13. Letvin NL, King NW , Reinherz EL, Hunt RD, Lane H, and Schlossman SF: T
    lymphocyte surface antigens in primates. Eur J Immunol 1983;13:345-347 .

14. Wood R, Dong H, Katzenstein DA, and Merigan TC: Quantification and
    comparison of HIV-1 proviral load in peripheral blood mononuclear cells and
    isolated CD4+ T cells. J Acquir Immune Def Synd 1993;6:237-240.

15. Oka S, Urayama K, Hirabayashi Y, Otmishi K, Goto H, Mitamura K, Kimura S,
    and Shimada K: Quantitative extimation of human immunodeficiency virus
    type-1 provirus in CD4+ T lymphocytes using the polymerase chain reaction.
    Mol Cell Probes 1991;5:137-142.

16. Aoki S, Yarchoan R, Thomas RV, Pluda, JM, Marczyk K, Broder S, and Mitsuya
    H: Quantitative analysis of HIV-1 proviral DNA in peripheral blood
    mononuclear cells from patients with AIDS or ARC: Decrease of proviral DNA
    content following treatment with 2_.3__ dideoxyinosine (ddI). AIDS Res Hum
    Retroviruses 1990;6:1331-1339.

17. Clark AGB, Ho1odniy M, Schwartz DH, Katzenstein DA, and Merigan TC: Decrease
    in HIV provirus in peripheral blood mononuclear cells during zidovudine and
    human rIL-2 administration. J Acquir Immune Defic Syndr 1992;5:52-59.

18. Dickover RE, Donovan RM, Goldstein E, Cohen SH, Bolton V, Huth RG, Liu GZ,
    and Carlson JR: Decreases in unintegrated HIV DNA are associated with
    antiretroviral therapy in AIDS patients. J Acquir Immune Defic Syndr
    1992;5:31-36.

19. Bush CE, Donovan RM, Smereck SM, Strang D, Markowitz N, and Saravolatz LD:
    Quantitation of unintegrated HIV-1 DNA in asymptomatic patients in the
    presence or absence of antiretroviral therapy. AIDS Res Hum Retroviruses
    1993;9:183-187.

20. Reimann KA, Burk1y LC, Burros B, Waite BCD, Lord CI, and Letvin NL: In vivo
    administration to rhesus monkeys of a CD4-specific monoclonal antibody
    capable of blocking AIDS virus replication. AIDS Res Hum Retroviruses
    1993;9:199-207.

<PAGE>
21. Burkly LC, Olson D, Shapiro R, Winkler G, Rosa JJ, Thomas DW, Williams C,
    and Chisholm P: Inhibition of HIV infection by a novel CD4 domain 2-specific
    monoclonal antibody. J Immunol 1992;149:1779-l787.

22. Moore JP , Sattentau QJ, Klasse PJ, and Burkly LC: A monoclonal antibody to
    CD4 domain 2 blocks soluble CD4-induced conformational changes in the
    envelope glycoproteins of human immunodeficiency virus type I (HIV-1) and
    HIV-1 infection of CD4+ cells. J Virol 1992;66:4784-4793.

23. Relmann KA, WaIte BCD, Lee-Parrtz DE, Lin W, Unchanska-Ziegler B, O Connell
    MJ, and Letvin NL: Use of human leukocyte-specific monoclonal antibodies for
    clinically imrnunophenotyping lymphocytes of rhesus monkeys. Cytometry
    1994;17:102-108.

24. Cate RL, Chick W, and Gilbert W: Comparison of the methylation patterns of
    the two rat insulin genes. J Biol Chem 1983;258:6645-6652.

25. Devlin PE, Ramachandran KL, and Cate RL: Southern analysis of genomic DNA
    with unique and degenerate oligonucleotide probes: A method for reducing
    probe degeneracy. DNA 1988;7:499-507.

26. Church GM and Gilbert W: Genomic sequencing. Proc Natl Acad Sci USA
    1984;81:1991-1995.

27. Trapp RG: BASIC AND CLINICAL BIOSTATISTICS. Appleton and Lange, Norwalk, CT
    1990.

28. Gowda SD, Stein BS, Mohagheghpour N, Benike CJ, and Engleman EG: Evidence
    that T cell activation is required for HIV-1 entry into CD4+ lymphocytes. J
    Immunol 1989;142:773-780.

29. Pantaleo G, Graziosi C, Demarest JF, Butini L, Montroni M, Fox CH, Orenstein
    JM, Kotler DP, and Fauci AS: HIV infection is active and progressive in
    lymphoid tissue during clinically latent stage of disease. Nature (London)
    1993;362:355-358.

30. Embertson J, Zupancic M, Ribas JL, Burke A, Racz P, Tenner-Racz K, and Haase
    AT: Massive covert infection of helper T lymphocytes and macrophages by HIV
    during the incubation period of AIDS. Nature (London) 1993;362:359-362.

                                                    Address reprint requests to:
                                                                KEITH A. REIMANN
                                                          HARVARD MEDICAL SCHOOL
                                                  DIVISION OF VIRAL PATHOGENESIS
                                                          DEPARTMENT OF MEDICINE
                                                     BETH ISRAEL HOSPITAL RE-113
                                                            330 BROOKLINE AVENUE
                                                                BOSTON, MA 02215

(1) Harvard Medical School, Beth Israel Hospital, Boston, Massachusetts 02215.
(2) Biogen, Inc., Cambridge, Massachusetts 02146.


FIG. 1. Magnitude of the antimouse Ig antibody responses correlated with CD4
cell count. The antimouse Ig antibody response in the plasma of SIVmac-infected
monkeys was quantitated using an ELISA technique. Data are expressed as
reciprocal endpoint dilutions. Samples were analyzed over a course of treatment
with either the anti-CD4 MAb 5A8 (_) or a control antibody (O). (A) Monkeys with
pretreatment CD4 counts of >800/microliters. (B) monkeys with pretreatment CD4
counts of <800/microliters.

<PAGE>
FIG. 2. CD4 cells were not cleared from circulation during the course of
treatment with anti-CD4 MAb. CD4+ lymphocytes were enumerated flow
cytometrica1ly in the peripheral blood of SIVmac-infected monkeys that received
repeated injections of anti-CD4 MAb. (A) Values from monkeys that had
pretreatment CD4 counts >800/microliters and were treated for 9 days. (B) Values
from monkeys that had pretreatment CD4 count of <800 and were treated for either
12 or 22 days. Shaded area represents period of MAb treatment

FIG. 3. Relative amount of SIVmac provirus is reproducibly determined using a
quantitative PCR technique. DNA from SIVmac-infected H9 cells was diluted into
rhesus monkey genomic DNA at the indicated ratios. This dilution also reflects
the SIVmac provirus/CD4 gene ratio since infected H9 cells contain, on average
2 SIVmac provirus copies per cell as determined by Southern blot analysis (data
not shown). These DNA were amplified by PCR, transferred to nylon membranes, and
hybridized with radiolabeled oligomers as described in Materials and Methods.
(A) Autoradiographs from 4 replicate assays. (B) The dilution of SIVmac
provirus in input DNA is plotted against the ratio of SIV/CD4 PCR products (mean
+/- SD).

FIG. 4. SIVmac PCR products and reference gene (CD4) products were determined
before and after anti-CD4 MAb treatment. SIVmac and CD4 PCR products from the
provirus assay described above are visualized after hybridization with a
radiolabeled probe. DNA was extracted from PBMC at various time points over the
treatment course with a control MAb (MOPC) or an anti-CD4 MAb (5A8).

FIG. 5. SIVmac provirus levels were unchanged or increased when control MAb was
administered but decreased when the anti-CD4 MAb 5A8 was given. SIVmac provirus
in PBMC of infected monkeys was quantified as described in Materials and Methods
over the course of treatment with either a control MAb or the anti-CD4 MAb 5A8
(mean +/- SD). Shaded area represents the time during which MAb was
administered for each individual monkey. (A) Changes in provirus level in 4
monkeys that received control MAb. (B) Change in provirus level in 3 monkeys
with CD4 count >800 CD4 cell/ microliters treated with anti-CD4 MAb. These
monkeys rapidly developed antimouse Ig responses and were treated for only 9
days. (C) Change in provirus level in 3 monkeys with <800 CD4 cells/microliters
treated with anti-CD4 MAb. These monkeys had a diminished humoral response to
mouse Ig and were treated for 12-22 days.

FIG. 6. Mean SIVmac provirus level decreased in monkeys that received the
anti-CD4 MAb 5A8. Mean SIVmac provirus levels (+/- SD) in infected monkeys
treated with the anti-CD4 MAb 548 (n = 6) or a control MAb (n = 4) where
posttreatment mean values are compared with pretreatment mean. The asterisk (*)
signifies significant decrease (Newman-Keuls multiple range test, p < 0.05).

TABLE 1. CHANGES IN CD4 LYMPHOCYTE AND MONOCYTE COUNT DURING MONOCLONAL ANTIBODY
TREATMENT


<PAGE>


                             TREATMENT(B)       BEFORE TREATMENT       DURING TREATMENT
              WEEKS
MONKEY NO.  INFECTED(C)  ANTIBODY  DURATION   CD4CELLS(C)  MONOCYTES(D)   CD4 CELLS

                                                           
  163          108          5A8       12d         402          108        860+/-143
  395           37          5A8        9d         831          104       2387+/-562
  133           37          5A8        9d        1067          270       1898+/-535
  152          148          5A8       22d       81+/-38     232+/-118      83+/-47
   81           56          5A8       22d      103+/-48     282+/-247      47+/-16
  284           56          5A8        9d     1458+/-107    348+/-60     1858+/-1149
  107           37         MOPC        9d         148          448        200+/-19
  410           37         MOPC        9d        1109          287       1361+/-165
  177          148         MOPC       22d      322+/-35     123+/-124     252+/-58
  239           56         MOPC        9d     1944+/-537    298+/-265    2195+/-214


<PAGE>
(a) Monkeys were infected with SIVmac the indicated number of weeks prior to
    treatment.
(b) Monkeys received intravenous injection of either a CD4-specific monoclonal
    antibody (5A8) or a control antibody (MOPC) at 3 mg/kg every 3 days for the
    indicated duration.
(c) CD4+ lymphocytes were enumerated by immunophenotyping and are expressed as
    mean cells/microliter +/- SD.
(d) Monocytes were enumerated by CBC and manual differential count and are
    expressed as mean cells/microliter +/- SD.